diff --git a/examples/LongitudinalLiNGAM.ipynb b/examples/LongitudinalLiNGAM.ipynb
index 2bb5243..8408a1f 100644
--- a/examples/LongitudinalLiNGAM.ipynb
+++ b/examples/LongitudinalLiNGAM.ipynb
@@ -24,7 +24,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "['1.24.4', '2.0.3', '0.20.1', '1.8.3']\n"
+      "['1.24.4', '2.0.3', '0.20.3', '1.9.1']\n"
      ]
     }
    ],
@@ -173,7 +173,7 @@
        "</svg>\n"
       ],
       "text/plain": [
-       "<graphviz.graphs.Digraph at 0x7f341f313a00>"
+       "<graphviz.graphs.Digraph at 0x7fd3e070e6a0>"
       ]
      },
      "execution_count": 3,
@@ -287,7 +287,7 @@
        "</svg>\n"
       ],
       "text/plain": [
-       "<graphviz.graphs.Digraph at 0x7f3478092ca0>"
+       "<graphviz.graphs.Digraph at 0x7fd3e070e100>"
       ]
      },
      "execution_count": 4,
@@ -394,7 +394,7 @@
        "</svg>\n"
       ],
       "text/plain": [
-       "<graphviz.graphs.Digraph at 0x7f341f347400>"
+       "<graphviz.graphs.Digraph at 0x7fd3e069ffd0>"
       ]
      },
      "execution_count": 5,
@@ -580,7 +580,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 504x216 with 2 Axes>"
       ]
@@ -592,7 +592,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAeoAAADQCAYAAADI+yJFAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuNSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/xnp5ZAAAACXBIWXMAAAsTAAALEwEAmpwYAAA3I0lEQVR4nO3dd3gUVffA8e9NKKFIldARkCIdJAiCgIpKEYj0DgqIqIivjWJ//dmwINKLSFGaovQIIqG3EBVBelEpBgkQWoBAkvP74655AyQkkOzObnI+z7NPdmdmZ8+uHM/M3Dv3GhFBKaWUUt7Jz+kAlFJKKZU8LdRKKaWUF9NCrZRSSnkxLdRKKaWUF9NCrZRSSnkxLdRKKaWUF9NCrRIYY5oaY+a7Yb/ZjTG7jTGF0nvfSqmUaW77Ni3UmYgx5k9jzEVjzHljTJQxZokxpmSiTd4DPnRtG2iMmWWM+dsYc8YYs94YU/cG+37FGPO7MeacMeYPY8wr/64TkRjgS2CIu76bUpmZm3P7AWPMSte2fyZep7ntGVqoM59WIpIbKAr8A4wCMMbUAfKKyCbXdrmBLUBtoAAwDVhijMmdzH4N0BPIDzQDBhhjOidaPxPoZYzJns7fRylluSu3o7HF+JVk1mtuu5kW6kxKRC4Bc4HKrkXNgdWJ1h8UkeEiEiEicSIyEcgGVExmfx+JyC8iEisie4AFQINE648AUUA993wjpRS4JbfDROQr4GAy6zW33UwLdSZljMkJdAL+PcquBuy5wfY1scm8PxX7NkBDYMc1q3YBNW4hXKVUKrkzt29Ac9uNsjgdgPK4+caYWCAXEAk0dS3PB5xL6g3GmDzAV8B/ReRMKj7jbexB4JRrlp9zfY5SKv15IreTo7ntRnpGnfk8JiL5gABgALDaGFMEe+nqtms3NsbkABYBm0Tkg5R2bowZgG2rftTV0SSx24DTaYpeKZUct+Z2CjS33UgLdSblapv6HogD7gO2ARUSb+PqHDIfOAI8ldI+jTG9sb0/m7jara5VCfgtbZErpW7EHbmdCprbbqSFOpMyVjC2l/YuIARonGh9VmyHlItALxGJv+b9pY0xYowp7XrdDXgfeFhErut0Yowpju1huunadUqp9OOG3PYzxgQAWV27DzDGZEu0vea2m2kbdeazyBgTBwjwFzZRdwC47pOsKyKbgfpAS2wyn7b9wwBoLiJrgZKu9x91LX8XKAhsSbTt1yLS3/W8KzAticvhSqn04a7cbgSsTPQ5F7G9yO93vdbcdjMjIk7HoLyEMeYR4BkReSwV274ORIrIhFRsmx17WayRiBxPc6BKqZuiue3btFArpZRSXszRNmpjzJfGmOPGmN+TWW+MMSONMfuNMduMMXd7Okal1M3RvFYqfTndmWwqdrjJ5DQHyrse/YBxHohJKZU2U9G8VirdOFqoRWQNcOoGmwQD08XaBOQzxhT1THRKqVuhea1U+vL2Xt/FgcOJXh9xLYtIvJExph/2yJxcuXLVvuuuuzwWoFLpQQQOH4bISICfT4hIRp42MFV5DZrbyvddugT790NMzK3ntbcX6lRxDSo/ESAoKEjCw8Mdjkip1Dt1Cjp0gF9+gUGD4KOPzF9Ox+QtNLeVL1u+3OZ2njwQGXnree10G3VKjmLv6ftXCf53b59SPm/vXqhXD9auhalTYdgwpyPyCM1rleGNHQvNm0OpUrBlS9r25e2FeiHQ09VLtB5wRkSuuzymlC9asQLq1oWoKAgNhV69nI7IYzSvVYYVGwsDBsCzz9pCvX493HFH2vbp6KVvY8ws7Og2txtjjgBvYYepQ0TGY4e+a4Gdfu0C8IQzkSqVvsaPt8l8112waBGUKeN0ROlH81plVqdPQ8eO9pL3K6/ABx+Av3/a9+tooRaRLimsF+BZD4WjlNvFxsKLL8KoUdCiBcyaZduvMhLNa5UZ7d8PLVvCwYMweTL07p1++84QncmU8gWnT0OnTvDjj7ZYf/RR+hxtK6WctXIltGsHfn7w00/QqFH67t/b26iVyhD274d777Vt0ZMmwaefapFWKiOYNAkeeQSKFoWwsPQv0qBn1Eq53erV0Latfb58Odx/v6PhKKXSQVwcvPwyjBgBzZrB7NmQN697PkvPqJVyo8mT4aGHIDAQNm/WIq1URnD2LLRqZYv0f/5jO4S6q0iDFmql3CIuDl56Cfr2hSZNYONGKFfO6aiUUml18KBtxlq+3N698dlnkMXN16b10rdS6ezsWejSBUJCYOBA2x7t7kRWSrnf2rW2GSsuDpYtgwcf9Mzn6hm1Uunojz+gfn2bxOPGweefa5FWKiOYMsVeHStY0DZjeapIgxZqpdLNunVwzz1w9Kgt1P37Ox2RUiqt4uLs4CW9e9s+Jps2Qfnyno1BC7VS6WDqVHuEnT+/Pdpu0sTpiJRSaXXuHLRpA598YocEDQmBfPk8H4cWaqXSIC4OBg+GJ56w909u3gwVKjgdlVIqrf76Cxo0sMV59Gj7cKoZS1vPlLpF589Dt26wcKG9zD1yJGTN6nRUSqm02rDBnknHxMAPP8DDDzsbj55RK3ULDh2yR9uLF9txu8eO1SKtVEbw1VfwwAN2DP5Nm5wv0qCFWqmbtnEj1KkDf/5pL4sNGADGOB2VUiot4uPh1VehZ097EL55s53dzhtooVbqJsyYYY+2c+e2R9tNmzodkVIqraKjoX17Oy3lk0/auzYKFHA6qv/RQq1UKsTHw+uvQ/fuUK+eHXy/UiWno1JKpdXhw3DffbBggR0SdMIE72vG0s5kSqUgOtpeDvv+e+jTx7ZHZ8vmdFRKqbQKC4PgYJvjixdD8+ZOR5Q0PaNW6gaOHLG3Xc2fD8OH2ynttEgr5ftmz4bGjSFHDtvvxFuLNGihVipZW7bYkcb27bO3YL3wgnYaU8rXxcfDW2/Z8fjr1LFn1VWqOB3VjWmhVioJc+bYM+ns2e09lY8+6nRESqm0unABOneGd96xgxT99BPcfrvTUaVMC7VSiYjA22/bZA4KskfbVas6HZVSKq2OHrUH33Pn2iFBJ0/2nWYs7UymlMvFi/D44/DNN9Crl+39mT2701EppdLq55+hdWs7Be3ChdCypdMR3Rw9o1YK+Ptv27Hk229h2DA7pZ0WaaV839y50LChveVqwwbfK9KghVopfvnFdhrbuRPmzYNBg7TTmFK+TgTefRc6dIBatWwzVrVqTkd1a7RQq0ztu+/sYAd+frB+vb2nUinl2y5etBPmvPEG9OgBK1ZAYKDTUd06LdQqUxKB996zwwbWqGGPtmvUcDoqpVRaHTtmh/mdNcsOCTptGgQEOB1V2mhnMpXpXLoEffvacbu7dYMvvvD9RFZKwdatttPYyZN2JME2bZyOKH3oGbXKVP75xx5tz5hh26+++kqLtFJOi4+PZ/To0dSoUYPixYvTvXt39u3bd1P7mD/fznolAuvWZZwiDVqoVSby2292JKJt22xP0Nde005jSnmDwYMHM2PGDEaPHs2GDRuoUqUKjRo14vDhwym+V8TeqdG2rR3zICzMdh7LSBwt1MaYZsaYPcaY/caYIUmsf9wYE2mM2ep69HUiTuWbTpw4wapVqzhw4AALF9qj7fh4WLsW2rVzOrqMTXNbpVZkZCRffPEFixcvpmHDhtxxxx0MHTqUrl27MmrUqBu+NybGjn0wZAh06gSrVkHRoh4J26McK9TGGH9gDNAcqAx0McZUTmLTOSJS0/X4wqNBKp8kIrz22muUK1eO119/gxo1ZhAcHE/FinFs2QJ33+10hBmb5rZKjoiwc+dOfv31V+Li4gDYuXMnVapUoWDBgldt+8gjj7B169Zk93X8ODz4IEyfbocEnTnTTrCRETnZmeweYL+IHAQwxswGgoGdDsakMoApU6awdOlStm/fyxtvBLJ+PZQpE0bFihMoWnSy0+FlBprb6jo7duygW7dunD59mhw5cnDx4kW++OILypcvz+7du7l48SI5ElXaX375hTJlyiS5r+3boVUrW6y/+cbeK52ROXnpuziQuAHiiGvZtdoZY7YZY+YaY0omtSNjTD9jTLgxJjwyMtIdsSofMmnSJAYN+pguXQKZNs2O3f3LLxVZsmQuZ8+edTq8zEBzW13l8uXLtGjRgueee46DBw+ya9cuJk+eTOfOnfH396dJkyb06dOH48ePEx8fz8KFC/nss88YMGDAdftavBjq14crV2DNmoxfpMH7O5MtAkqLSHVgOTAtqY1EZKKIBIlIUKFChTwaoPI+f/9dgJdeuo+ff7Zzzr71FuTNm4eAgAAt1N5DczsTCQkJoUyZMvTp0wc/P1t2mjRpQseOHZk+fTpTpkwhb968lCtXjnz58vHmm28ya9YsqiUaSkwEPv3U3n5VsaLtNBYU5NQ38iwnL30fBRIfRZdwLUsgIicTvfwC+MgDcSkftmQJRER8R/bsl1izJht16tjloaGh5MuXj2LFijkbYOagua2ucvz4ccqWLXvd8jvvvJNDhw6RM2dOxo0bx4gRI7hw4QL58uXDJLol4/JlePpp+PJLO0jRtGmQM6cnv4GznDyj3gKUN8aUMcZkAzoDCxNvYIxJ3H+vNbDLg/EpHyICw4fbdqsKFQy5czdh6tRnCQkJYdiwYXTt2pXhw4cnHM0rt9LcVldp2LAhP/zwA+fPn09YFh8fz9y5c2nUqFHCsuzZs5M/f/6rivSJE/Dww7ZIv/mmnSs+MxVpcPCMWkRijTEDgGWAP/CliOwwxrwDhIvIQmCgMaY1EAucAh53Kl7lvS5fhmeftSOMtW0L06dn5+zZhYwZM4aRI0dSsmRJli1bRs2aNZ0ONVPQ3FbXqlSpEsHBwTz44IO88sor5MqVi7FjxxIQEEDwDQbY37nTHnwfPWp7dXfp4sGgvYgREadjSFdBQUESHh7udBjKQ06etPdEr15tBzB55x07wYYvM8b8LCKZpPUt9TS3fVt8fDwzZsxg5syZxMTE0Lp1a5566qmrenontnSpvTc6Rw5YsADq1vVwwOksLXmtY30rn7Vrlz3aPnIEvv7ajtutlPJOfn5+9OjRgx49etxwOxEYNQpeeMFOS7loEZRM8p6AzMPHzz1UZrVsGdSrB+fOwcqVWqSVygiuXLGdxp5/3vbuXrdOizRooVY+5t+j7RYtoHRpe4vGvfc6HZVSKq1OnYJmzWDCBBg61M4Vnzu301F5B730rXzGlSv2SHvcOHu0PWOGJrJSGcGePdCyJRw6ZIcETeHqeKajZ9TKJ0RFQfPmtkgPHgzz5mmRVspXnTx5kl27dhETE8Py5baj2JkzEBqqRTopekatvN7evbbT2B9/wNSp0KuX0xEppW5FdHQ0Tz/9NAsXLiQwMJC//w7m4sVhVK7sx6JFtjlLXU/PqJVXW7HCHm2fOmWPtrVIK+W7nnnmGeLj41m+fCUXL35MdPTHxMcvAepz5co+p8PzWlqoldcaPx6aNoXixW2nsfvuczoipdStOnXqFAsWLODNN4fTqNE5jhwJ5sUXY5ky5TSxsVE8/PDDXLhwwekwvZIWauV1YmNh4EB7m0bTprBhAyQz251SymGrV6/m6aefpm/fvixatIj4+Pgkt9u9ezexsaW5664oLl2qR4MGXzJ48CmqV69CtmzZqFq1Kt99952Ho/cNN2yjNsa8eKP1IjI8fcNRmd2xY5do3DiCvXvLULjwDO677yjZsg0EApwOLUPR3FZpFR8fT9u2bfnxxx8pUaIE1atXZ+jQocybN4/JkydfM6nGZbp0mUh0dCjZs2elZcvxlCr1J02afEarVq2oX78+BQoU4M8//3TuC3mxlM6ob3M9goCnsXPKFgf6A3e7NzSVEUVHRzNr1izGjRvH3r17r1q3b59Qrtxx9u0rycsv72H27OJs2LCW1q1bJ3uUrm6Z5rZKk65du7JkyRJGjhzJp59+SnR0NIGBgaxdu5a1a9dete3Agds5fPgLihWzE+aEhw+nY8eOnDlzhnHjxvHSSy+xbNkyateu7dC38XIikuIDWAPcluj1bcCa1LzX04/atWuL8k7r16+XwMBAefTRR6V3795SqFAhGTRokMTHx8uqVSK33XZZ/P2j5KefYhPeExsbK9WqVZPly5c7GLlnYSeu8Ei+aG6rW/H7779L3rx5pWfPngnLYmNjpU6dOtKpUycZPHiwa5nIf/4jAiJly+6R06dFQkJCpECBApI/f3656667pHv37tK1a1dp0KCBxMbGJveRPi8teZ3aNurCwOVEry+7limVKrGxsXTq1Ikvv/ySxYsXM3nyZPbs2cOCBQv4z3+289BDEBBwhj59JtGkiX/C+/z9/WnZsiUbN250MPoMTXNb3bTly5dz1113cfbs2YRl/v7+tG/fnj179nDbbbdx9qy9rXLECGjRYi8lSjxN3rzQvHlz/vrrL55++mn++usvQkNDueOOO/jhhx/w9/dP/kMzsdTeRz0dCDPGzHO9fgyY5paIVIa0bt06ihQpwqOPPpqwLE+e/JQoMZuRI6vTtCm0bv0jy5atA1656r27du2iVatWHo4409DcVilau3YtU6dO5fTp05w+fZpNmzYRGxvLli1bGDJkCB9++CEAP//8M3v37qVBgx7ce68dA2H8eHjiidLUqnWMQYMGMWjQIMC2W5cpU4atW7eSNWtWJ7+e10vVGbWIvAc8AUS5Hk+IyPvuDExlLFeuXCEg4H8dws6etcOArlhRk/Llf2DxYujZszXh4eFMnjyZuLg44uLimDJlCps3b6ZDhw4ORp9xaW6rlIwYMYLu3btTrVo1zp8/z7Zt26hevTqFChWie/fufPTRR1SvXp06derw7bff0q/fV3ToUIqICDt5zlNPQbZs2VixYgXHjh2jVKlS3HHHHRw/fpzQ0FAt0qmR2mvkwH3YJAYoBJS51evt7nxoO5Z3io6OloIFC0p4eLgcPChSpYqIv3+8lC79gcycOTNhu23btkmdOnUkMDBQAgMDJSgoSH777TcHI/c8PNhGLZrb6gb27t0ruXPnll27dsm5c+ckb968cuzYMXnggQfk//7v/6Rq1apy++23S/bs2eX222+X5577RbJmFalQQWTv3qT3GR8fL/Hx8Z79Il4gLXmdqkvfxpi3sL1DKwJTgKzA10CD9DxoUBlXzpw5mTRpEg8++CaXL88B/ClRYgBBQWfp2PF/l7qrVatGWFgYhw4dQkS44447nAs6E9DcVkmJiYlhwIABzJw5Ez8/P+rXr0/fvn3JnTs3hQsXpmfPnixfvpxt27axcOFCBg58gfbt9zN8uB8PPQTffAP58ye978S3banUSW1nsjZAayAaQET+xvYOVSrVzp5tw6VLi8mVK4auXUcycWIn5syZk2QHkn8vjym309xW1xk0aBCRkZHMmTOH8uXL8+uvv/LDDz8kXPo+duwY+fLlwxjDtm1/EBc3l+HD/XjmGQgJSb5Iq1uT2s5kl0VEjDECYIzJ5caYVAYTFwevvgoffQRNmhi+/bYg+fMPdjosZWluq6tcunSJadOmsXv3bgoVKsQLL7xAaGgon3zyCf3796dly5acO3eOiRMn8tprExk27GGgMqNHw7PPOh19xpTaQv2NMWYCkM8Y8yTQG/jCfWGpjOL8eejWDRYuhP79YeRI0L4jXkVzW13l/PnzGGMoXLgwxhgWLFhAmzZtMMYQERFBlixZKFmyJC+88A3Hjo0jLi4LWbI8yvTpJ6lW7WMaNWrk9FfIcFLb6/sTYC7wHbYt600RGenOwJTvO3QIGjSAxYth1CgYO1aLtLfR3FbXKliwIIGBgaxevRqAypUrs2vXLh5++GHq1q3L33//zauv7iIiYiZZslygUqXeFCu2g6xZs9KmTRt27Njh8DfIgFLT4wwYlppl3vDQnqHeYcMGkcBAkTx5RJYudToa34JnRybT3FbXmTdvnhQpUkTGjRsnGzdulP/+979SqFAh2bbtdxk6VARE/PxWSYUK9SQkJET2798vn3zyieTMmVM6derkdPheKS15ndrOZA8nsax52g8TVEY0YwY88ADkzg2bNtkZsJTX0txW13nssceYO3cuoaGhDBw4kKNHj/Ljj+t5660qfPABNG9+BGOasXTpTJo3b86dd97JSy+9RLt27a4b51ulXUqzZz0NPAOUNcZsS7TqNmC9OwNTvic+Ht58E957Dxo3hu++g4IFnY5KJUVzW6WkQYMGNGhg79I7fNgOULRtmx0StGbNgyxbdpm8efNe9R5jDLGxsQ5Em7Gl1JlsJvAD8AEwJNHycyJyym1RKZ8THQ09e8L330OfPrY9Ols2p6NSN6C5rVIlLAyCg22OL14MzZvDqVNVyZIlCy1btmTEiBHccccdzJo1i3nz5lG/fn2nQ85wblioReQMcAboAmCMCcRODJzbGJNbRA65P0Tl7Y4csYm8dSsMHw7/+Q/omAbeTXNbpcbs2dCrVxzZsp2iceNPOHOmFnFxHShQoADdu3dn8+bNdO3alaioKKpWrUpAQACvvvqq02FnOKlqozbGtDLG7AP+AFYDf2KPxlUmt2UL3HMP7Ntnb8F64QUt0r5Ec1slJT4e3noLunQBP7+fGTLkex57rDyfffYZHTt2JD4+njFjxvDggw9y4sQJjDEcPXqUsWPH6u1Z7pCaHmfAb0BB4FfX6weAybfag82dD+0Z6j5AwiMyMlLmzBEJCBApXVpk+3ano8s48Gyvb83tDCQ+Pl5+/fVXWbVqlURHR1+17tKlS7JkyRL55ptv5MSJE8nuIzpapEMHERDJlu1rOXTon4R1MTExUqNGDVmyZEnCsgsXLsixY8ckLi4u/b9QBpKWvE5tr+8rInIS8DPG+InISuz4wGlijGlmjNljjNlvjBmSxPrsxpg5rvWbjTGl0/qZ6uYNGjTouvF5CxUaRadOEBRk27CqVnUoOJVWmtsZxIEDB6hduzbt27dnyJAhlCxZkqlTpwKwceNGypQpw4cffshXX31FuXLlGD9+/HX7OHoUGjWCuXOhQ4fNtGsXQsmSgQnrs2XLRvfu3Vm6dGnCshw5clC4cGH8/FJbTtRNS001B34CcgOjgFnA58CGWz06cO3THzgAlAWyYY/sK1+zzTPAeNfzzsCclParR93pD9dZ9KJFi+TCBZHHHrskIAJT5NIlp6PLePDsGbXmdgYQHx8vVatWlREjRiTMTLVjxw4pWrSorF27VooUKSKLFy9O2P7gwYNStGhR+eWXXxKWhYeLFCsmkitXnDRq9LFky5ZN/P39pXfv3hIZGZmw3SuvvCKvv/66575cBpGWvE7tIVAwcBF4AVjqSsJWN3NAkIR7gP0iclBELgOzXZ9z7ef+O4n9XKCJ0alXHOHn58fdd7ekcWNYsCA7ZctOAJ5g5MiPnQ5NpY3mdgawadMm4uPjGThwYMLVr8qVKzNw4EDee+89KlasyKOPPpqwfZkyZXjqqaf46quvAHsG3bAh+PvHU7BgMA0bnubAgQPkyZOHNWvWULlyZUJCQti2bRtTpkyhR48ejnzPzCq1Q4hGi0gckBNYhJ0GT9L42cWBw4leH3EtS3IbEYnF9lK97s5cY0w/Y0y4MSY8MjIyjWGppAQE1Oeee2DnTpg3D/r3PwvAoUPaOdiXaW5nDAcOHODSpUu0aNGCfv36sXXrVgBKlixJVFQUBQoUuO49+fPn5/z5aN59Fzp0gFq1YPDg76hcOZZ3332X5cuXAxAREcGZM2do27YtQUFBDB8+nAoVKnjy62V6qe31/ZQx5hiwDQgHfnb99QoiMlFEgkQkqFChQk6HkwG15cKFZfj5wfr19lasQYMGATBq1CiHY1Npobnt+44cOcLQoUM5cuQInTp14s4776Rp06YsXLiQGTNm0LZtW0JDQ686qL58+TJTpsxm9+7XeeMN6NEDVqyAo0d/pUGDBpw+fZoXX3yRTZs2ERUVRatWrejUqROVKlUiT548Dn7bzCm1s2e9DFQVkRPp+NlHgZKJXpdwLUtqmyPGmCxAXuBkOsagbkAE3n8f7HwNGzl8uA01a/6TsD6peaSVz9Hc9nEffvghXbp0ITAwkHfeeYdBgwYxYMAAunTpQpUqVRgwYADZs2enfv36PPPMM+TLl48JExZw+PBUoqJK8sEHMHiwva2ycuXKfP3111SqVIn69etToUIFRIRDhw7x9ttvU6dOHRYtWkRw8LUtGcqdUttGfQC4kM6fvQUob4wpY4zJhu1QsvCabRYCvVzP2wOhrkZ55WaXLtmj7Ndft9NUdu48Efhfka5ataoOFZgxaG77uNDQULp168bLL7/MxIkTWbNmDRs3biRLlixMmzaNnDlz8vzzz/P9999z7NgxfvzxOH//PY+YmPJ8/z0MGfK/sQ/at2/PH3/8wezZszl9+jQRERE899xzxMXF0bRpU6KjowkICHD2C2dGqelxBtQCtgITgJH/Pm61B1ui/bYA9mL/Z/Gaa9k7QGvX8wDgW2A/EAaUTWmf2jM07Y4dE6lXTwRE3n1XxNWJVHkInu31rbnt4+67776renSLiJw9e1by5Mkjp06dSlh2+fJlGTcuQnLmjJcSJUQSdfi+ytGjR6Vz584CSM6cOaVv375y4sQJiYiIkFKlSsn69evd+XUyrLTktZFUHMQaY8KAdcB2ID5RkZ+W7JscEhQUJOHhXtPE5nN++w1atYKTJ2H6dGjXzumIMh9jzM8ikuZ7mVP5WZrbPm7atGmMGDGCpUuXUrhwYS5fvszzzz9PVFQUs2fPBmDq1GkMHHiEc+eG4uf3Cx07zmTy5HfJmTNnsvtds2YN7du3p3bt2hQoUIClS5fyyiuvMGTIdbfFq1RIS16nto06q4i8eCsfoHzHwoXQtSvkywdr18LddzsdkfIAzW0f17NnT/bt20fFihWpUaMGe/fupVatWsycOROAJUt+4tlnc3Hhwmt07gyffnonL70UwTPPPJMwIEpSGjVqxB9//EFISAjnz59n2LBhlChRwkPfSiWW2jPq97FjAC8CYv5dLl44y44edd88Efj4Y9tWFRQECxZA0aJOR5V5efiMWnM7gzh58iTbtm2jRIkSlC9fHoDjx+Guu3YRFVWJd96xfU6MgTNnzlC6dGn2799PQZ2L1iM8cUbdxfV3aKJlgh15SPmwmBjo3x+mToVOnWDKFMiRw+molAdpbmcQBQsW5IEHHkh4vX27bcY6fbo077+/n6FDyyWsy5s3L4ULF+aff/7RQu0DUlWoRaSMuwNRnhcZCW3bwrp18Pbb8OabOvNVZqO57XtEhM2bNxMREUGdOnUSLkefOXOGFStWkDVrVi5ffoTHH89OnjzQps0ILl68iO3LZ+3evZtTp05Rtqwej/mCGxZqY8yDIhJqjGmb1HoR+d49YSl3+/13e7R97Jidc7ZTJ6cjUp6kue2bjhw5QnBwMBcuXKB8+fL06dOHPn36UL16dQYOHEi9eveyb18rDhzISvnyZ1m5Mg8xMZ2oX78+/v7+tGnThj179jB06FDeeustvdXKR6R0Rt0YCCXpsX8F0GT2QUuW2Hlmc+eGNWugTh2nI1IO0Nz2QY8//jjBwcG88cYbGGOIioqiQYMGjB8/nrVrNzNqVGWWLoXGjSPZvv1ubrttB8WLl2XdunUJA6MUK1aMTz75hMcee8zpr6NSKzX3cAFlUrPMGx56r2Xy4uNFPv1UxBiRWrVEDh92OiKVFDx7H7Xmto84dOiQ3H777RITE3PV8k6dOkmxYtWlUSMREHnjDZG4OJHg4GD56quvHIpWXSsteZ3akcm+S2LZ3LQfJihPuXwZ+vWDl16CNm3s7Vd6p4VCc9tnnDt3jrx585ItW7Zrlpfkn38WsHkzzJwJ77wDfn5QoEABzp8/71C0Kj2l1EZ9F1AFyHtNW1Ye7MhCygecPGkHLlm9Gl577X+JrDIvzW3fU7FiRWJjY1m9ejWNGzcGYOlS+PHHt4mPP8+yZZdo3Nj+p4uMjGTRokW88cYbToas0klKbdQVgZZAPq5uyzoHPOmmmFQ62rXLdho7cgS+/tqO260Umts+x9/fn9GjR9OhQweefLIfhw4F8/XXdxMQcJgWLUbz3HNrePLJJ7l06RJjx45lwIABlCmjnfozghsWahFZACwwxtwrIhs9FJNKJ8uWQceOEBAAK1fCvfc6HZHyFprbvqlly5ZMmjSVF1/MysGDdahe/SDLlpUgMHAkISEhLFiwgKxZszJ9+nQaNmzodLgqnaR2wJM2xpgdwEVgKVAdeEFEvnZbZOqWicDo0fCf/0C1anZo0FKlnI5KeSnNbR8RERFBmzZ92Lr1VWJi7iMgYDhPPJGFIkUGAraIt2zZ0uEolTuktqXyERE5i71U9idQDnjFXUGpW3flCjz7LAwcCC1b2sFMtEirG9Dc9hHBwYPYu/crRBowfTrs2PEYn3/+GcuXL3c6NOVmqS3UWV1/HwW+FZEzbopHpUFUFDRvDuPG2Yng582z90ordQOa2z5g6tSjhIePJkuWAoSGGnr0gLJly/Lqq68yadIkp8NTbpbaS9+LjDG7sZfHnjbGFAIuuS8sdbP27rWdxv74w47b3auX0xEpH6G57eXGjoWBA4uRPfsBwsLyUrr0/9aVLFmSkydPOhab8oxUnVGLyBCgPhAkIleAC0CwOwNTqbdiBdStC6dOQWioFmmVeprb3is2FgYMsE1ZjzwST86cD3Hx4q6rtpk1a9ZVE3GojOmGhdoYMyjRyyYiEgcgItHAQHcGplJnwgRo2hSKF4ewMLjvPqcjUr5Ac9u7nT4NLVrAmDHw8suwaJE/w4a9ziOPPMLo0aNZvHgx3bt3JywsjGeffdbpcJWbpXRG3TnR86HXrGuWzrGomxAbazuM9e9vC/WGDaC3TKqboLntpfbvh3r1YNUqmDzZzhXv7w99+/Zl5syZhIWFMXr0aKpUqcLGjRvJnz+/0yErN0upjdok8zyp18pDTp+2s139+CO8+CJ89JFNZKVugua2F1q50o4i6OcHy5cL0dE/0Lv3XESEdu3a8eijj+r90ZlQSmfUkszzpF4rD9i/3w5cEhoKkybBp59qkVa3RHPby0yaBI88AkWL2masuXMH8vLLL1O7dm3q1KnDkCFD9DJ3JpXSGXUNY8xZ7BF2DtdzXK91PGAP2Lt3L3///Tc1atRg27b8tHWNyrx8Odx/v6OhKd+mue0l4uJsO/SIEdCsmZ0f/uDBX5k/fz47duwgT548APTs2ZMqVaoQHh5OUFCQs0Erj0ppCFE9V3PIyZMn6dy5Mzt27ODOO+/k559rEhMzggoV/Fi0yFCunNMRKl+mue0dzp6Fzp3hhx/sSIIffwxZssCyZcvo0KFDQpEGyJ07Nx07dmTZsmVaqDMZnUPJS/Xu3ZuqVaty8OBf3HPPWi5eHEWOHBsZPHiBFmmlMoCDB20z1vLlMH48fPaZLdIAefLkITIy8rr3REZGXlW8VeaghdoLHTt2jLVr1zJkyPu0a5eV4cNtD+8vvjjGzJljnQ5PKZVGa9bAPfdARISdPOeppyAuLo6lS5cyfvx4ypUrR0hICBs2bEh4z+bNm1m0aBEdO3Z0MHLlhNSOTKY86PTp09x2W3WaNMnB7t12SND+/WHTphI6CpFSPm7KFFuYy5SBxYuhfHk4evQoTZs2JWfOnNSsWZOPPvqIChUq0Lp1a6pUqYIxhh07djBt2jQKFy7s9FdQHqaF2gsdO1aOI0e+I1euWJYty0KTJnb5zJkzefDBB50NTil1S+LioE+fSKZNK0SFCn/x3XdZKF++OAD9+/enXbt2/Pe//wXgypUrtG3blqZNm1K3bl1EhAceeIAcOXI4+RWUQ7RQe5lp06BfvywULpyNK1cas3VrWy5erMj8+fMJDQ296lKYUso3nDsHd9+9i/37K1Gjxjpq1/6aRo2+YcyYMbRo0YJVq1YxZ86chO2zZs3K66+/Tu/evXn77bedC1x5BW2j9hLx8TBkCDz+ODRsCDt23EZIyAh27tzJmDFjKFWqFGFhYRQpUsTpUJVSN+Gvv6BWrWj27y/PsGHn2br1PiZPHs+aNWt45plniIyMxBhDtmzZrnpfrly5iImJcShq5U0cOaM2xhQA5gClsXPgdhSRqCS2iwO2u14eEpHWnorRk86fh+7dYcEC2xY9ciRkzQp16tShTp06ToenVKppbl9twwZo0wbOnPGjV69vGDSoa8K6qlWrcv/997N582aqVavG9OnT6d27NwAiwujRo2ndOkP+LOomOXXpewiwQkQ+NMYMcb0enMR2F0Wkpkcj87BDh6B1a9i+HUaNsjPlGB3AUfkuzW2Xr76Cvn2hZElo0WIkd955Jdltx4wZQ7NmzVi9ejU1atRg2bJlREREsHLlSg9GrLyVU5e+g4FprufTgMccisNRGzfaWzT++ANCQuyUdlqklY/L9LkdHw+vvgo9e0L9+rB5M/Tr14hJkyZx4sSJhO22b9/OqlWraNGiBTVr1uT333+nZs2aHDp0iG7duhEWFkbBggUd/CbKWzh1Rl1YRCJcz48Byd1vEGCMCQdigQ9FZH5SGxlj+gH9AEqVKpXOobrHjBnQp4+dnnLlSqhUyemIlEoXmTq3o6OhRw+YNw+efNJOU5k1K9x777307NmTatWq0bFjR86dO8f8+fOZMGFCwuxXt99+Oy+88ILD30B5IyPinvH3jTE/AUn1fHoNmCYi+RJtGyUi183VZowpLiJHjTFlgVDsvLkHbvS5QUFBEh4enrbg3Sg+Ht58E957Dxo3hu++Az1oVokZY34WEa8dI1JzO2mHD9tmrG3bSBik6NorZDt37mTx4sUEBATQvn17ihUr5kywyuPSktduO6MWkYeSW2eM+ccYU1REIowxRYHjyezjqOvvQWPMKqAWcMNk9mbR0fZy2Pff27arMWPgmo6eSnk9ze3rhYVBcLDN8cWLoXnzpLerXLkylStX9mxwyuc51Ua9EOjlet4LWHDtBsaY/MaY7K7ntwMNgJ0eizCdHTkCjRrB/Pn2aHviRC3SKkPKdLk9e7a9OpYjh+13klyRVupWOVWoPwQeNsbsAx5yvcYYE2SM+cK1TSUg3BjzG7AS247lk8m8ZYvtNLZvHyxcCC+8oJ3GVIaVaXI7Ph7eegu6dIE6dexZdZUqTkelMiJHOpOJyEmgSRLLw4G+rucbgGoeDi3dffMN9OoFRYrAjz9C1apOR6SU+2SW3L5wwQ5O9O238MQTdjz+7NmdjkplVDoymZuIwH//C506QVCQPdrWIq2U7zt61DZjzZ1r54+ePFmLtHIvHevbDS5etEfZc+bYs+kJEzSRlcoIfv7Z9uw+e9aOJNiqldMRqcxAz6jTWUSE7VjyzTcwbJid0k6LtFK+b+5cOw5/liywfr0WaeU5WqjT0a+/2k4lO3faAQ8GDdJOY0r5OhF4913o0AFq1rTNWNWrOx2Vyky0UN+iM2fOcOHChYTX338P990Hfn72aDs42MHglFLp4uJF6NYN3njDTpwTGgqFkxtrTSk30UJ9k3799VcaNmxI8eLFCQwMpGPHTgwdep527exRdlgY1KjhdJRKqbQ6dgweeABmzYL334fp0yEgwOmoVGakhfom/PPPPzRr1ozevXsTFRXFwYN/s337ID78MDfdugkrV9rbsJRSvm3rVjv2wfbt9mrZ0KHajKWco4X6JkyZMoXg4GCeeOIJTp3KSnBwHnbvrk1g4Ej69VunR9tKZQDz50ODBrZtet06O5+0Uk7SQn0TDh48yN13381vv9lOY9u22Z6gjzyyhT///MPp8JRSaSACH34IbdvaMQ/CwqBWLaejUkoL9U2pVasW06ZF0aCBHT5w7Vpo1eoyq1atombNmk6Hp5S6RTExdsyDoUPtIEWrVkHRok5HpZSlA56kkgicONGbTZuyUrjwIWbPvkJ8fBRt275N3bp1qa73ayjlk44ft5e3N2yAd96B11/X9mjlXbRQp0JMDPTvD1OnZic4+CKBgZ/Stes8cubMSffu3Rk0aJDTISqlbsH27XbgkuPH7SBFHTo4HZFS19NCnYLISNtmtW4dvP02vPlmDoz5HPjc6dCUUmmweLGd+SpPHlizxo7Jr5Q30jbqG/j9d3uLRni4nXP2rbf0kphSvk4EPv3UjtldoYLtNKZFWnkzLdTJCAmB+vXtZe81a2wHE6WUb7t8Gfr2hZdfhnbtbIfQ4sWdjkqpG9NCfQ0R+Owz225Vrpw92q5Tx+molFJpdeIEPPwwfPmlHRJ0zhzImdPpqJRKmbZRJ3L5MgwYAJMm2Xbp6dMhVy6no1JKpdXOnfbg++hRmDEDunZ1OiKlUk/PqF1OnoRHHrFF+rXX4NtvtUgrlREsXQr33gvR0bB6tRZp5Xu0UAO7dkHdurBpE3z9tZ3Szk9/GaV8mgiMHAmPPgplythmrLp1nY5KqZuX6cvRsmVQrx6cOwcrV9op7ZRSvu3KFXj6aXj+edu7e906KFXK6aiUujWZtlCLwKhR0KKFPdressVeHlNK+bZTp6BZM5gwwQ4J+t13kDu301EpdesyZWeyK1fskfa4cfZoe8YMTWSlMoI9e6BlSzh0yHYG7dHD6YiUSrtMV6ijouwwgStWwODBdkJ4bY9WyvctX25zO1s2CA21U1UqlRFkqhK1d69tj16zBqZOtVPaaZFWyveNHQvNm0PJkrbTmBZplZFkmjK1YoXt8XnqlD3a7tXL6YiUUmkVG2vHPnj2WVuoN2yA0qWdjkqp9JUpCvWECdC0qR0qMCwM7rvP6YiUUml1+rTtDDpmjB0SdP58uO02p6NSKv1l6Dbq2Fh46SV7L2WLFjBrlp0pRynl2/bvt53GDh6EyZOhd2+nI1LKfTJsoT5zxk6ksWwZvPgifPQR+Ps7HZVSKq1WrrQTavj52Q5kjRs7HZFS7uXIpW9jTAdjzA5jTLwxJtkJ5owxzYwxe4wx+40xQ1K7/wMH7D3RK1bYIUE//VSLtFKe4O7cnjTJDvVbpAhs3qxFWmUOTrVR/w60BdYkt4Exxh8YAzQHKgNdjDGVU9rxuXN2Dul//rFH2337plfISqlUcFtuHz4M/fpBkyawcSPceWd6hayUd3OkUIvILhHZk8Jm9wD7ReSgiFwGZgPBKe17714IDLRH2/ffnw7BKqVSzZ25ffy4Haho8WLImzc9olXKN3hzG3Vx4HCi10eAJIfUN8b0A/q5Xsbs3m1+L1/ezdGlr9uBE04HcRM0Xveq6HQAbnbLuf355+b3zz93c3Tpx9f+3Wm87nXLee22Qm2M+QkoksSq10RkQXp+lohMBCa6PjdcRJJtG/NGvhazxutexphwp2O4Ec3t1NF43csX473V97qtUIvIQ2ncxVGgZKLXJVzLlFIO0txWyrO8ecCTLUB5Y0wZY0w2oDOw0OGYlFJpp7mt1E1w6vasNsaYI8C9wBJjzDLX8mLGmBAAEYkFBgDLgF3ANyKyIxW7n+imsN3J12LWeN3L1+JNoLl9FY3XvTJNvEZE0jMQpZRSSqUjb770rZRSSmV6WqiVUkopL+bzhdrdQxamN2NMAWPMcmPMPtff/MlsF2eM2ep6eLyjTUq/lzEmuzFmjmv9ZmNMaU/HeE08KcX7uDEmMtFv6uiYdcaYL40xx40xvyez3hhjRrq+zzZjzN2ejtFpmttui1Nz203cltci4tMPoBL2RvJVQFAy2/gDB4CyQDbgN6CyQ/F+BAxxPR8CDEtmu/MO/qYp/l7AM8B41/POwBwvj/dxYLRTMSYRcyPgbuD3ZNa3AH4ADFAP2Ox0zA78Rprb6R+j5rZ743VLXvv8GbW4cchCNwkGprmeTwMecyiOG0nN75X4e8wFmhhjjAdjTMyb/vumioisAU7dYJNgYLpYm4B8xpiinonOO2huu4Xmthu5K699vlCnUlJDFhZ3KJbCIhLhen4MKJzMdgHGmHBjzCZjzGOeCS1Ban6vhG3E3m5zBijokeiul9r/vu1cl5vmGmNKJrHem3jTv1lv5k2/k+Z2+stouX1L/169eazvBMaDQxamhxvFm/iFiIgxJrn74+4QkaPGmLJAqDFmu4gcSO9YM5FFwCwRiTHGPIU9Y3jQ4ZgyPc1tze10kOFz2ycKtfjYkIU3itcY848xpqiIRLgueRxPZh9HXX8PGmNWAbWwbTWekJrf699tjhhjsgB5gZOeCe86KcYrIolj+wLbnujNMsUwm5rbmtspyGi5fUv/XjPLpW9vGrJwIdDL9bwXcN1ZgzEmvzEmu+v57UADYKfHIkzd75X4e7QHQsXVW8IBKcZ7TTtQa+yIWN5sIdDT1Uu0HnAm0WVV9T+a2zdHc9tZt5bXTveSS4dedm2w1/ljgH+AZa7lxYCQa3rb7cUeub7mYLwFgRXAPuAnoIBreRDwhet5fWA7tofjdqCPA3Fe93sB7wCtXc8DgG+B/UAYUNbhfwcpxfsBsMP1m64E7nI43llABHDF9e+3D9Af6O9ab4Axru+znWR6PWfkh+a22+LU3HZfrG7Jax1CVCmllPJimeXSt1JKKeWTtFArpZRSXkwLtVJKKeXFtFArpZRSXkwLtVJKKeXFfGLAE+V+xph/by0BO/JSHBDpen2P2HF2lVI+RnPb9+ntWeo6xpi3sTP8fJJoWRax4/4qpXyU5rZv0jNqlSxjzFTgEnaIw/XGmLMkSnLXnKstReRPY0x3YCB2KrrNwDMiEudM5EqpG9Hc9i3aRq1SUgKoLyIvJreBMaYS0AloICI1sZfWunkmPKXULdLc9hF6Rq1S8m0qjp6bALWBLa5pa3OQzIQESimvobntI7RQq5REJ3oey9VXYQJcfw0wTUSGeiwqpVRaaW77CL30rW7Gn8DdAMaYu4EyruUrgPbGmEDXugLGmDsciVApdSv+RHPba2mhVjfjO6CAMWYHMAA7ow0ishN4HfjRGLMNWA4UTXYvSilvo7ntxfT2LKWUUsqL6Rm1Ukop5cW0UCullFJeTAu1Ukop5cW0UCullFJeTAu1Ukop5cW0UCullFJeTAu1Ukop5cX+HwJl09d1rhaVAAAAAElFTkSuQmCC",
       "text/plain": [
        "<Figure size 504x216 with 2 Axes>"
       ]
@@ -773,10 +773,10 @@
       "x3(2) <--- x0(1) (b>0) (100.0%)\n",
       "x3(2) <--- x2(2) (b<0) (100.0%)\n",
       "x2(2) <--- x3(1) (b>0) (100.0%)\n",
-      "x2(2) <--- x1(1) (b>0) (100.0%)\n",
       "x4(2) <--- x3(1) (b>0) (100.0%)\n",
       "x1(2) <--- x3(1) (b>0) (100.0%)\n",
-      "x1(2) <--- x2(1) (b>0) (100.0%)\n"
+      "x1(2) <--- x2(1) (b>0) (100.0%)\n",
+      "x1(2) <--- x0(1) (b<0) (100.0%)\n"
      ]
     }
    ],
@@ -1088,7 +1088,7 @@
       " [0.06 0.   1.   0.04 1.  ]\n",
       " [0.8  0.   0.   0.   0.07]\n",
       " [0.03 0.02 1.   0.   0.1 ]\n",
-      " [0.91 0.   0.   0.01 0.  ]]\n",
+      " [0.91 0.   0.01 0.01 0.  ]]\n",
       "B(2,1):\n",
       "[[1.   1.   0.91 1.   0.92]\n",
       " [1.   0.86 1.   1.   0.96]\n",
@@ -1165,21 +1165,21 @@
        "      <th>1</th>\n",
        "      <td>x4(1)</td>\n",
        "      <td>x4(2)</td>\n",
-       "      <td>-0.278507</td>\n",
+       "      <td>-0.201297</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>2</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x4(2)</td>\n",
-       "      <td>0.185780</td>\n",
+       "      <td>0.319831</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>3</th>\n",
        "      <td>x1(1)</td>\n",
        "      <td>x4(2)</td>\n",
-       "      <td>0.351397</td>\n",
+       "      <td>0.373447</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
@@ -1191,234 +1191,234 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>5</th>\n",
-       "      <td>x3(1)</td>\n",
+       "      <td>x4(1)</td>\n",
        "      <td>x3(2)</td>\n",
-       "      <td>-0.161284</td>\n",
+       "      <td>-0.299875</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>6</th>\n",
+       "      <td>x3(1)</td>\n",
+       "      <td>x3(2)</td>\n",
+       "      <td>-0.331575</td>\n",
+       "      <td>1.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
        "      <td>x2(1)</td>\n",
        "      <td>x3(2)</td>\n",
        "      <td>0.495256</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>7</th>\n",
+       "      <th>8</th>\n",
        "      <td>x1(1)</td>\n",
        "      <td>x3(2)</td>\n",
-       "      <td>-0.579338</td>\n",
+       "      <td>-0.294184</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>8</th>\n",
+       "      <th>9</th>\n",
        "      <td>x0(1)</td>\n",
        "      <td>x3(2)</td>\n",
        "      <td>0.186140</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>9</th>\n",
+       "      <th>10</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x2(2)</td>\n",
-       "      <td>0.400577</td>\n",
+       "      <td>0.543159</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>10</th>\n",
+       "      <th>11</th>\n",
        "      <td>x1(1)</td>\n",
        "      <td>x2(2)</td>\n",
-       "      <td>0.326661</td>\n",
+       "      <td>0.352668</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>11</th>\n",
+       "      <th>12</th>\n",
        "      <td>x0(1)</td>\n",
        "      <td>x2(2)</td>\n",
        "      <td>0.161875</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>12</th>\n",
+       "      <th>13</th>\n",
        "      <td>x2(2)</td>\n",
        "      <td>x1(2)</td>\n",
        "      <td>-0.692908</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>13</th>\n",
+       "      <th>14</th>\n",
+       "      <td>x4(1)</td>\n",
+       "      <td>x1(2)</td>\n",
+       "      <td>0.299131</td>\n",
+       "      <td>1.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>15</th>\n",
        "      <td>x0(1)</td>\n",
        "      <td>x1(2)</td>\n",
        "      <td>-0.563879</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>14</th>\n",
+       "      <th>16</th>\n",
        "      <td>x4(2)</td>\n",
        "      <td>x1(2)</td>\n",
        "      <td>0.476373</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>15</th>\n",
-       "      <td>x3(1)</td>\n",
-       "      <td>x0(2)</td>\n",
-       "      <td>-0.495518</td>\n",
-       "      <td>1.00</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>16</th>\n",
+       "      <th>17</th>\n",
        "      <td>x4(1)</td>\n",
        "      <td>x0(1)</td>\n",
        "      <td>-0.586968</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>17</th>\n",
+       "      <th>18</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x1(1)</td>\n",
        "      <td>0.388875</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>18</th>\n",
-       "      <td>x0(1)</td>\n",
-       "      <td>x0(2)</td>\n",
-       "      <td>0.202197</td>\n",
+       "      <th>19</th>\n",
+       "      <td>x1(1)</td>\n",
+       "      <td>x2(1)</td>\n",
+       "      <td>0.357268</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>19</th>\n",
-       "      <td>x1(1)</td>\n",
+       "      <th>20</th>\n",
+       "      <td>x3(1)</td>\n",
        "      <td>x0(2)</td>\n",
-       "      <td>0.191862</td>\n",
+       "      <td>-0.387450</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>20</th>\n",
+       "      <th>21</th>\n",
        "      <td>x1(1)</td>\n",
        "      <td>x4(1)</td>\n",
        "      <td>-0.356674</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>21</th>\n",
+       "      <th>22</th>\n",
+       "      <td>x0(1)</td>\n",
+       "      <td>x0(2)</td>\n",
+       "      <td>0.202197</td>\n",
+       "      <td>1.00</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>23</th>\n",
        "      <td>x1(1)</td>\n",
-       "      <td>x2(1)</td>\n",
-       "      <td>0.357268</td>\n",
+       "      <td>x0(2)</td>\n",
+       "      <td>0.258939</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>22</th>\n",
+       "      <th>24</th>\n",
        "      <td>x1(1)</td>\n",
        "      <td>x1(2)</td>\n",
-       "      <td>-0.100172</td>\n",
+       "      <td>-0.181191</td>\n",
        "      <td>0.99</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>23</th>\n",
+       "      <th>25</th>\n",
+       "      <td>x4(1)</td>\n",
+       "      <td>x2(2)</td>\n",
+       "      <td>-0.115873</td>\n",
+       "      <td>0.99</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>26</th>\n",
        "      <td>x2(1)</td>\n",
        "      <td>x1(2)</td>\n",
-       "      <td>0.169769</td>\n",
+       "      <td>0.163286</td>\n",
        "      <td>0.99</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>24</th>\n",
+       "      <th>27</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x4(1)</td>\n",
        "      <td>-0.108293</td>\n",
        "      <td>0.98</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>25</th>\n",
-       "      <td>x4(1)</td>\n",
-       "      <td>x3(2)</td>\n",
-       "      <td>-0.158863</td>\n",
-       "      <td>0.98</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>26</th>\n",
-       "      <td>x2(1)</td>\n",
-       "      <td>x2(2)</td>\n",
-       "      <td>-0.064596</td>\n",
-       "      <td>0.97</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>27</th>\n",
+       "      <th>28</th>\n",
        "      <td>x0(1)</td>\n",
        "      <td>x4(2)</td>\n",
        "      <td>-0.146124</td>\n",
        "      <td>0.97</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>28</th>\n",
+       "      <th>29</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x0(1)</td>\n",
        "      <td>0.080405</td>\n",
        "      <td>0.97</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>29</th>\n",
-       "      <td>x3(1)</td>\n",
-       "      <td>x2(1)</td>\n",
-       "      <td>0.032170</td>\n",
-       "      <td>0.94</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
        "      <th>30</th>\n",
-       "      <td>x2(1)</td>\n",
-       "      <td>x4(2)</td>\n",
-       "      <td>-0.099157</td>\n",
-       "      <td>0.94</td>\n",
+       "      <td>x4(1)</td>\n",
+       "      <td>x0(2)</td>\n",
+       "      <td>-0.130167</td>\n",
+       "      <td>0.97</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>31</th>\n",
-       "      <td>x3(1)</td>\n",
-       "      <td>x1(2)</td>\n",
-       "      <td>0.079244</td>\n",
-       "      <td>0.93</td>\n",
+       "      <td>x2(1)</td>\n",
+       "      <td>x4(2)</td>\n",
+       "      <td>-0.092469</td>\n",
+       "      <td>0.97</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>32</th>\n",
-       "      <td>x4(1)</td>\n",
-       "      <td>x0(2)</td>\n",
-       "      <td>-0.005440</td>\n",
-       "      <td>0.92</td>\n",
+       "      <td>x2(1)</td>\n",
+       "      <td>x2(2)</td>\n",
+       "      <td>-0.068454</td>\n",
+       "      <td>0.95</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>33</th>\n",
-       "      <td>x0(2)</td>\n",
-       "      <td>x4(2)</td>\n",
-       "      <td>0.261939</td>\n",
-       "      <td>0.91</td>\n",
+       "      <td>x3(1)</td>\n",
+       "      <td>x2(1)</td>\n",
+       "      <td>0.032170</td>\n",
+       "      <td>0.94</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>34</th>\n",
        "      <td>x2(1)</td>\n",
        "      <td>x0(2)</td>\n",
-       "      <td>0.019144</td>\n",
-       "      <td>0.91</td>\n",
+       "      <td>0.023336</td>\n",
+       "      <td>0.92</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>35</th>\n",
-       "      <td>x0(2)</td>\n",
+       "      <td>x3(1)</td>\n",
        "      <td>x1(2)</td>\n",
-       "      <td>-0.029275</td>\n",
-       "      <td>0.90</td>\n",
+       "      <td>0.000911</td>\n",
+       "      <td>0.92</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>36</th>\n",
-       "      <td>x4(1)</td>\n",
-       "      <td>x1(2)</td>\n",
-       "      <td>-0.014277</td>\n",
-       "      <td>0.90</td>\n",
+       "      <td>x0(2)</td>\n",
+       "      <td>x4(2)</td>\n",
+       "      <td>0.261939</td>\n",
+       "      <td>0.91</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>37</th>\n",
-       "      <td>x4(1)</td>\n",
-       "      <td>x2(2)</td>\n",
-       "      <td>-0.019646</td>\n",
-       "      <td>0.85</td>\n",
+       "      <td>x0(2)</td>\n",
+       "      <td>x1(2)</td>\n",
+       "      <td>-0.029275</td>\n",
+       "      <td>0.90</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>38</th>\n",
@@ -1478,17 +1478,17 @@
        "    </tr>\n",
        "    <tr>\n",
        "      <th>46</th>\n",
-       "      <td>x1(2)</td>\n",
-       "      <td>x3(2)</td>\n",
-       "      <td>-0.174134</td>\n",
+       "      <td>x2(2)</td>\n",
+       "      <td>x4(2)</td>\n",
+       "      <td>0.084880</td>\n",
        "      <td>0.02</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>47</th>\n",
-       "      <td>x2(2)</td>\n",
-       "      <td>x4(2)</td>\n",
-       "      <td>0.045734</td>\n",
-       "      <td>0.01</td>\n",
+       "      <td>x1(2)</td>\n",
+       "      <td>x3(2)</td>\n",
+       "      <td>-0.174134</td>\n",
+       "      <td>0.02</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>48</th>\n",
@@ -1504,43 +1504,43 @@
       "text/plain": [
        "     from     to    effect  probability\n",
        "0   x1(1)  x0(1)  0.257084         1.00\n",
-       "1   x4(1)  x4(2) -0.278507         1.00\n",
-       "2   x3(1)  x4(2)  0.185780         1.00\n",
-       "3   x1(1)  x4(2)  0.351397         1.00\n",
+       "1   x4(1)  x4(2) -0.201297         1.00\n",
+       "2   x3(1)  x4(2)  0.319831         1.00\n",
+       "3   x1(1)  x4(2)  0.373447         1.00\n",
        "4   x2(2)  x3(2) -0.428210         1.00\n",
-       "5   x3(1)  x3(2) -0.161284         1.00\n",
-       "6   x2(1)  x3(2)  0.495256         1.00\n",
-       "7   x1(1)  x3(2) -0.579338         1.00\n",
-       "8   x0(1)  x3(2)  0.186140         1.00\n",
-       "9   x3(1)  x2(2)  0.400577         1.00\n",
-       "10  x1(1)  x2(2)  0.326661         1.00\n",
-       "11  x0(1)  x2(2)  0.161875         1.00\n",
-       "12  x2(2)  x1(2) -0.692908         1.00\n",
-       "13  x0(1)  x1(2) -0.563879         1.00\n",
-       "14  x4(2)  x1(2)  0.476373         1.00\n",
-       "15  x3(1)  x0(2) -0.495518         1.00\n",
-       "16  x4(1)  x0(1) -0.586968         1.00\n",
-       "17  x3(1)  x1(1)  0.388875         1.00\n",
-       "18  x0(1)  x0(2)  0.202197         1.00\n",
-       "19  x1(1)  x0(2)  0.191862         1.00\n",
-       "20  x1(1)  x4(1) -0.356674         1.00\n",
-       "21  x1(1)  x2(1)  0.357268         1.00\n",
-       "22  x1(1)  x1(2) -0.100172         0.99\n",
-       "23  x2(1)  x1(2)  0.169769         0.99\n",
-       "24  x3(1)  x4(1) -0.108293         0.98\n",
-       "25  x4(1)  x3(2) -0.158863         0.98\n",
-       "26  x2(1)  x2(2) -0.064596         0.97\n",
-       "27  x0(1)  x4(2) -0.146124         0.97\n",
-       "28  x3(1)  x0(1)  0.080405         0.97\n",
-       "29  x3(1)  x2(1)  0.032170         0.94\n",
-       "30  x2(1)  x4(2) -0.099157         0.94\n",
-       "31  x3(1)  x1(2)  0.079244         0.93\n",
-       "32  x4(1)  x0(2) -0.005440         0.92\n",
-       "33  x0(2)  x4(2)  0.261939         0.91\n",
-       "34  x2(1)  x0(2)  0.019144         0.91\n",
-       "35  x0(2)  x1(2) -0.029275         0.90\n",
-       "36  x4(1)  x1(2) -0.014277         0.90\n",
-       "37  x4(1)  x2(2) -0.019646         0.85\n",
+       "5   x4(1)  x3(2) -0.299875         1.00\n",
+       "6   x3(1)  x3(2) -0.331575         1.00\n",
+       "7   x2(1)  x3(2)  0.495256         1.00\n",
+       "8   x1(1)  x3(2) -0.294184         1.00\n",
+       "9   x0(1)  x3(2)  0.186140         1.00\n",
+       "10  x3(1)  x2(2)  0.543159         1.00\n",
+       "11  x1(1)  x2(2)  0.352668         1.00\n",
+       "12  x0(1)  x2(2)  0.161875         1.00\n",
+       "13  x2(2)  x1(2) -0.692908         1.00\n",
+       "14  x4(1)  x1(2)  0.299131         1.00\n",
+       "15  x0(1)  x1(2) -0.563879         1.00\n",
+       "16  x4(2)  x1(2)  0.476373         1.00\n",
+       "17  x4(1)  x0(1) -0.586968         1.00\n",
+       "18  x3(1)  x1(1)  0.388875         1.00\n",
+       "19  x1(1)  x2(1)  0.357268         1.00\n",
+       "20  x3(1)  x0(2) -0.387450         1.00\n",
+       "21  x1(1)  x4(1) -0.356674         1.00\n",
+       "22  x0(1)  x0(2)  0.202197         1.00\n",
+       "23  x1(1)  x0(2)  0.258939         1.00\n",
+       "24  x1(1)  x1(2) -0.181191         0.99\n",
+       "25  x4(1)  x2(2) -0.115873         0.99\n",
+       "26  x2(1)  x1(2)  0.163286         0.99\n",
+       "27  x3(1)  x4(1) -0.108293         0.98\n",
+       "28  x0(1)  x4(2) -0.146124         0.97\n",
+       "29  x3(1)  x0(1)  0.080405         0.97\n",
+       "30  x4(1)  x0(2) -0.130167         0.97\n",
+       "31  x2(1)  x4(2) -0.092469         0.97\n",
+       "32  x2(1)  x2(2) -0.068454         0.95\n",
+       "33  x3(1)  x2(1)  0.032170         0.94\n",
+       "34  x2(1)  x0(2)  0.023336         0.92\n",
+       "35  x3(1)  x1(2)  0.000911         0.92\n",
+       "36  x0(2)  x4(2)  0.261939         0.91\n",
+       "37  x0(2)  x1(2) -0.029275         0.90\n",
        "38  x0(2)  x3(2) -0.106739         0.84\n",
        "39  x0(2)  x2(2)  0.250640         0.80\n",
        "40  x4(1)  x2(1) -0.169832         0.24\n",
@@ -1549,8 +1549,8 @@
        "43  x2(1)  x4(1) -0.171814         0.11\n",
        "44  x4(2)  x2(2)  0.155502         0.07\n",
        "45  x3(2)  x1(2) -0.155433         0.05\n",
-       "46  x1(2)  x3(2) -0.174134         0.02\n",
-       "47  x2(2)  x4(2)  0.045734         0.01\n",
+       "46  x2(2)  x4(2)  0.084880         0.02\n",
+       "47  x1(2)  x3(2) -0.174134         0.02\n",
        "48  x3(2)  x4(2) -0.146344         0.01"
       ]
      },
@@ -1611,38 +1611,38 @@
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>6</th>\n",
+       "      <th>10</th>\n",
+       "      <td>x3(1)</td>\n",
+       "      <td>x2(2)</td>\n",
+       "      <td>0.543159</td>\n",
+       "      <td>1.0</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>7</th>\n",
        "      <td>x2(1)</td>\n",
        "      <td>x3(2)</td>\n",
        "      <td>0.495256</td>\n",
        "      <td>1.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>14</th>\n",
+       "      <th>16</th>\n",
        "      <td>x4(2)</td>\n",
        "      <td>x1(2)</td>\n",
        "      <td>0.476373</td>\n",
        "      <td>1.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>9</th>\n",
-       "      <td>x3(1)</td>\n",
-       "      <td>x2(2)</td>\n",
-       "      <td>0.400577</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>17</th>\n",
+       "      <th>18</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x1(1)</td>\n",
        "      <td>0.388875</td>\n",
        "      <td>1.0</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>21</th>\n",
+       "      <th>3</th>\n",
        "      <td>x1(1)</td>\n",
-       "      <td>x2(1)</td>\n",
-       "      <td>0.357268</td>\n",
+       "      <td>x4(2)</td>\n",
+       "      <td>0.373447</td>\n",
        "      <td>1.0</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
@@ -1651,11 +1651,11 @@
       ],
       "text/plain": [
        "     from     to    effect  probability\n",
-       "6   x2(1)  x3(2)  0.495256          1.0\n",
-       "14  x4(2)  x1(2)  0.476373          1.0\n",
-       "9   x3(1)  x2(2)  0.400577          1.0\n",
-       "17  x3(1)  x1(1)  0.388875          1.0\n",
-       "21  x1(1)  x2(1)  0.357268          1.0"
+       "10  x3(1)  x2(2)  0.543159          1.0\n",
+       "7   x2(1)  x3(2)  0.495256          1.0\n",
+       "16  x4(2)  x1(2)  0.476373          1.0\n",
+       "18  x3(1)  x1(1)  0.388875          1.0\n",
+       "3   x1(1)  x4(2)  0.373447          1.0"
       ]
      },
      "execution_count": 24,
@@ -1708,39 +1708,39 @@
        "  </thead>\n",
        "  <tbody>\n",
        "    <tr>\n",
-       "      <th>15</th>\n",
+       "      <th>20</th>\n",
        "      <td>x3(1)</td>\n",
        "      <td>x0(2)</td>\n",
-       "      <td>-0.495518</td>\n",
+       "      <td>-0.387450</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>18</th>\n",
+       "      <th>22</th>\n",
        "      <td>x0(1)</td>\n",
        "      <td>x0(2)</td>\n",
        "      <td>0.202197</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>19</th>\n",
+       "      <th>23</th>\n",
        "      <td>x1(1)</td>\n",
        "      <td>x0(2)</td>\n",
-       "      <td>0.191862</td>\n",
+       "      <td>0.258939</td>\n",
        "      <td>1.00</td>\n",
        "    </tr>\n",
        "    <tr>\n",
-       "      <th>32</th>\n",
+       "      <th>30</th>\n",
        "      <td>x4(1)</td>\n",
        "      <td>x0(2)</td>\n",
-       "      <td>-0.005440</td>\n",
-       "      <td>0.92</td>\n",
+       "      <td>-0.130167</td>\n",
+       "      <td>0.97</td>\n",
        "    </tr>\n",
        "    <tr>\n",
        "      <th>34</th>\n",
        "      <td>x2(1)</td>\n",
        "      <td>x0(2)</td>\n",
-       "      <td>0.019144</td>\n",
-       "      <td>0.91</td>\n",
+       "      <td>0.023336</td>\n",
+       "      <td>0.92</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table>\n",
@@ -1748,11 +1748,11 @@
       ],
       "text/plain": [
        "     from     to    effect  probability\n",
-       "15  x3(1)  x0(2) -0.495518         1.00\n",
-       "18  x0(1)  x0(2)  0.202197         1.00\n",
-       "19  x1(1)  x0(2)  0.191862         1.00\n",
-       "32  x4(1)  x0(2) -0.005440         0.92\n",
-       "34  x2(1)  x0(2)  0.019144         0.91"
+       "20  x3(1)  x0(2) -0.387450         1.00\n",
+       "22  x0(1)  x0(2)  0.202197         1.00\n",
+       "23  x1(1)  x0(2)  0.258939         1.00\n",
+       "30  x4(1)  x0(2) -0.130167         0.97\n",
+       "34  x2(1)  x0(2)  0.023336         0.92"
       ]
      },
      "execution_count": 25,
@@ -1791,7 +1791,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -1812,6 +1812,158 @@
     "to_index = 12 # index of x2(2). (index:2)+(n_features:5)*(timepoint:2) = 12\n",
     "plt.hist(result.total_effects_[:, to_index, from_index])"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Prior knowledge\n",
+    "\n",
+    "### Prior knowledge matrix\n",
+    "\n",
+    "The shape of the prior knowledge is the same as for LongitudinalLiNGAM.adjacency_matrices_.\n",
+    "\n",
+    "The elements of prior knowledge matrix are defined as follows:\n",
+    "* ``0`` : :math:`x_i` does not have a directed path to :math:`x_j`\n",
+    "* ``1`` : :math:`x_i` has a directed path to :math:`x_j`\n",
+    "* ``-1`` : No prior knowledge is available to know if either of the two cases above (0 or 1) is true.\n",
+    "\n",
+    "### Example\n",
+    "\n",
+    "In this example, the path from x2(2) to x1(2) in graph B(2, 2) is prohibited."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/svg+xml": [
+       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
+       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
+       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
+       "<!-- Generated by graphviz version 2.43.0 (0)\n",
+       " -->\n",
+       "<!-- Title: %3 Pages: 1 -->\n",
+       "<svg width=\"250pt\" height=\"305pt\"\n",
+       " viewBox=\"0.00 0.00 250.45 305.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
+       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 301)\">\n",
+       "<title>%3</title>\n",
+       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-301 246.45,-301 246.45,4 -4,4\"/>\n",
+       "<!-- x0(2) -->\n",
+       "<g id=\"node1\" class=\"node\">\n",
+       "<title>x0(2)</title>\n",
+       "<ellipse fill=\"none\" stroke=\"black\" cx=\"34.45\" cy=\"-279\" rx=\"34.39\" ry=\"18\"/>\n",
+       "<text text-anchor=\"middle\" x=\"34.45\" y=\"-275.3\" font-family=\"Times,serif\" font-size=\"14.00\">x0(2)</text>\n",
+       "</g>\n",
+       "<!-- x2(2) -->\n",
+       "<g id=\"node3\" class=\"node\">\n",
+       "<title>x2(2)</title>\n",
+       "<ellipse fill=\"none\" stroke=\"black\" cx=\"107.45\" cy=\"-18\" rx=\"34.39\" ry=\"18\"/>\n",
+       "<text text-anchor=\"middle\" x=\"107.45\" y=\"-14.3\" font-family=\"Times,serif\" font-size=\"14.00\">x2(2)</text>\n",
+       "</g>\n",
+       "<!-- x0(2)&#45;&gt;x2(2) -->\n",
+       "<g id=\"edge3\" class=\"edge\">\n",
+       "<title>x0(2)&#45;&gt;x2(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M32.29,-260.87C29.5,-234.67 26.04,-183.25 35.45,-141 44.7,-99.46 51.92,-89.46 75.45,-54 78.2,-49.85 81.45,-45.7 84.79,-41.79\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"87.57,-43.93 91.68,-34.16 82.38,-39.24 87.57,-43.93\"/>\n",
+       "<text text-anchor=\"middle\" x=\"51.45\" y=\"-144.8\" font-family=\"Times,serif\" font-size=\"14.00\">0.14</text>\n",
+       "</g>\n",
+       "<!-- x4(2) -->\n",
+       "<g id=\"node5\" class=\"node\">\n",
+       "<title>x4(2)</title>\n",
+       "<ellipse fill=\"none\" stroke=\"black\" cx=\"80.45\" cy=\"-192\" rx=\"34.39\" ry=\"18\"/>\n",
+       "<text text-anchor=\"middle\" x=\"80.45\" y=\"-188.3\" font-family=\"Times,serif\" font-size=\"14.00\">x4(2)</text>\n",
+       "</g>\n",
+       "<!-- x0(2)&#45;&gt;x4(2) -->\n",
+       "<g id=\"edge7\" class=\"edge\">\n",
+       "<title>x0(2)&#45;&gt;x4(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M43.54,-261.21C50.13,-249.02 59.17,-232.32 66.65,-218.49\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"69.85,-219.93 71.53,-209.47 63.7,-216.6 69.85,-219.93\"/>\n",
+       "<text text-anchor=\"middle\" x=\"75.45\" y=\"-231.8\" font-family=\"Times,serif\" font-size=\"14.00\">0.21</text>\n",
+       "</g>\n",
+       "<!-- x1(2) -->\n",
+       "<g id=\"node2\" class=\"node\">\n",
+       "<title>x1(2)</title>\n",
+       "<ellipse fill=\"none\" stroke=\"black\" cx=\"149.45\" cy=\"-105\" rx=\"34.39\" ry=\"18\"/>\n",
+       "<text text-anchor=\"middle\" x=\"149.45\" y=\"-101.3\" font-family=\"Times,serif\" font-size=\"14.00\">x1(2)</text>\n",
+       "</g>\n",
+       "<!-- x1(2)&#45;&gt;x2(2) -->\n",
+       "<g id=\"edge4\" class=\"edge\">\n",
+       "<title>x1(2)&#45;&gt;x2(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M141.15,-87.21C135.17,-75.1 126.98,-58.53 120.17,-44.76\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"123.3,-43.17 115.73,-35.76 117.02,-46.27 123.3,-43.17\"/>\n",
+       "<text text-anchor=\"middle\" x=\"148.95\" y=\"-57.8\" font-family=\"Times,serif\" font-size=\"14.00\">&#45;0.42</text>\n",
+       "</g>\n",
+       "<!-- x3(2) -->\n",
+       "<g id=\"node4\" class=\"node\">\n",
+       "<title>x3(2)</title>\n",
+       "<ellipse fill=\"none\" stroke=\"black\" cx=\"190.45\" cy=\"-192\" rx=\"34.39\" ry=\"18\"/>\n",
+       "<text text-anchor=\"middle\" x=\"190.45\" y=\"-188.3\" font-family=\"Times,serif\" font-size=\"14.00\">x3(2)</text>\n",
+       "</g>\n",
+       "<!-- x3(2)&#45;&gt;x1(2) -->\n",
+       "<g id=\"edge1\" class=\"edge\">\n",
+       "<title>x3(2)&#45;&gt;x1(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M182.3,-174.27C179.49,-168.49 176.33,-161.96 173.45,-156 169.64,-148.11 165.5,-139.51 161.76,-131.7\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"164.91,-130.18 157.43,-122.68 158.6,-133.21 164.91,-130.18\"/>\n",
+       "<text text-anchor=\"middle\" x=\"189.45\" y=\"-144.8\" font-family=\"Times,serif\" font-size=\"14.00\">0.25</text>\n",
+       "</g>\n",
+       "<!-- x3(2)&#45;&gt;x2(2) -->\n",
+       "<g id=\"edge5\" class=\"edge\">\n",
+       "<title>x3(2)&#45;&gt;x2(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M199.21,-174.36C201.74,-168.69 204.17,-162.22 205.45,-156 211.72,-125.43 204.64,-115.72 192.45,-87 185.65,-71 184.14,-65.88 171.45,-54 163.16,-46.25 152.84,-39.64 142.92,-34.32\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"144.5,-31.19 133.99,-29.8 141.33,-37.44 144.5,-31.19\"/>\n",
+       "<text text-anchor=\"middle\" x=\"223.95\" y=\"-101.3\" font-family=\"Times,serif\" font-size=\"14.00\">&#45;0.27</text>\n",
+       "</g>\n",
+       "<!-- x4(2)&#45;&gt;x1(2) -->\n",
+       "<g id=\"edge2\" class=\"edge\">\n",
+       "<title>x4(2)&#45;&gt;x1(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M93.43,-175.01C103.74,-162.31 118.32,-144.34 130.02,-129.93\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"132.94,-131.89 136.53,-121.92 127.51,-127.47 132.94,-131.89\"/>\n",
+       "<text text-anchor=\"middle\" x=\"134.45\" y=\"-144.8\" font-family=\"Times,serif\" font-size=\"14.00\">0.38</text>\n",
+       "</g>\n",
+       "<!-- x4(2)&#45;&gt;x2(2) -->\n",
+       "<g id=\"edge6\" class=\"edge\">\n",
+       "<title>x4(2)&#45;&gt;x2(2)</title>\n",
+       "<path fill=\"none\" stroke=\"black\" d=\"M76.54,-173.82C72.47,-152.99 67.56,-116.96 74.45,-87 77.89,-72.02 85.09,-56.51 91.89,-44.09\"/>\n",
+       "<polygon fill=\"black\" stroke=\"black\" points=\"95.02,-45.68 96.94,-35.26 88.94,-42.2 95.02,-45.68\"/>\n",
+       "<text text-anchor=\"middle\" x=\"90.45\" y=\"-101.3\" font-family=\"Times,serif\" font-size=\"14.00\">0.22</text>\n",
+       "</g>\n",
+       "</g>\n",
+       "</svg>\n"
+      ],
+      "text/plain": [
+       "<graphviz.graphs.Digraph at 0x7fd3e069fac0>"
+      ]
+     },
+     "execution_count": 27,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "pk = np.ones((3, 2, 5, 5)) * -1\n",
+    "\n",
+    "# T=2, tau=0\n",
+    "pk[2, 0, 1, 2] = 0\n",
+    "model = lingam.LongitudinalLiNGAM(n_lags=n_lags, prior_knowledge=pk)\n",
+    "model = model.fit(X_t)\n",
+    "\n",
+    "make_dot(model.adjacency_matrices_[2, 0], labels=[f'x{i}(2)' for i in range(5)])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {
+    "tags": []
+   },
+   "source": [
+    "Comparing with graph B(2, 2) above, we can confirm that the path from x2(2) to x1(2) has been eliminated."
+   ]
   }
  ],
  "metadata": {
@@ -1831,6 +1983,13 @@
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
    "version": "3.8.10"
+  },
+  "widgets": {
+   "application/vnd.jupyter.widget-state+json": {
+    "state": {},
+    "version_major": 2,
+    "version_minor": 0
+   }
   }
  },
  "nbformat": 4,
diff --git a/lingam/longitudinal_lingam.py b/lingam/longitudinal_lingam.py
index d3e1413..8a697c9 100644
--- a/lingam/longitudinal_lingam.py
+++ b/lingam/longitudinal_lingam.py
@@ -25,13 +25,21 @@ class LongitudinalLiNGAM:
        Workshop on Machine Learning for Signal Processing (MLSP2013), pp. 1--6, Southampton, United Kingdom, 2013.
     """
 
-    def __init__(self, n_lags=1, measure="pwling", random_state=None):
+    def __init__(self, n_lags=1, prior_knowledge=None, measure="pwling", random_state=None):
         """Construct a model.
 
         Parameters
         ----------
         n_lags : int, optional (default=1)
             Number of lags.
+        prior_knowledge : array-like, shape (T, n_lags + 1, n_features, n_features), optional (default=None)
+            Prior knowledge used for causal discovery, where ``n_features`` is the number of features.
+
+            The elements of prior knowledge matrix are defined as follows [1]_:
+
+            * ``0`` : :math:`x_i` does not have a directed path to :math:`x_j`
+            * ``1`` : :math:`x_i` has a directed path to :math:`x_j`
+            * ``-1`` : No prior knowledge is available to know if either of the two cases above (0 or 1) is true.
         measure : {'pwling', 'kernel'}, default='pwling'
             Measure to evaluate independence : 'pwling' or 'kernel'.
         random_state : int, optional (default=None)
@@ -43,6 +51,12 @@ def __init__(self, n_lags=1, measure="pwling", random_state=None):
         self._causal_orders = None
         self._adjacency_matrices = None
 
+        if prior_knowledge is not None:
+            prior_knowledge = check_array(prior_knowledge, ensure_2d=False, allow_nd=True)
+            if len(prior_knowledge.shape) != 4:
+                raise ValueError("prior_knowledge must be 4D.")
+        self._Aknw = prior_knowledge
+
     def fit(self, X_list):
         """Fit the model to datasets.
 
@@ -75,25 +89,106 @@ def fit(self, X_list):
                 raise ValueError("X_list must be a list with the same shape")
             X_t.append(X.T)
 
-        M_tau, N_t = self._compute_residuals(X_t)
-        B_t, causal_orders = self._estimate_instantaneous_effects(N_t)
-        B_tau = self._estimate_lagged_effects(B_t, M_tau)
+        n_taus = self._n_lags + 1
+
+        if self._Aknw is None:
+            M_tau, N_t = self._compute_residuals(X_t)
+            B_t, causal_orders = self._estimate_instantaneous_effects(N_t)
+            B_tau = self._estimate_lagged_effects(B_t, M_tau)
+
+            # output B(t,t), B(t,t-τ)
+            self._adjacency_matrices = np.empty(
+                (self._T, n_taus, self._p, self._p)
+            )
+            self._adjacency_matrices[:, :] = np.nan
+            for t in range(self._n_lags, self._T):
+                self._adjacency_matrices[t, 0] = B_t[t]
+                for l in range(self._n_lags):
+                    if t - l != 0:
+                        self._adjacency_matrices[t, l + 1] = B_tau[t, l]
+
+            self._residuals = np.zeros((self._T, self._n, self._p))
+            for t in range(self._T):
+                self._residuals[t] = N_t[t].T
+            self._causal_orders = causal_orders
+        else:
+            if (self._T, n_taus, self._p, self._p) != self._Aknw.shape:
+                raise ValueError(
+                    "The shape of prior knowledge must be (T, n_lags + 1, n_features, n_features)"
+                )
+
+            X_t = np.vstack(X_t)
+
+            # estimate only instantaneous and lag effects
+            pk = np.zeros((self._T * self._p, self._T * self._p))
+            for t in range(self._T):
+                col_end = (t + 1) * self._p
+                col_start = max(col_end - self._p * n_taus, 0)
+                pk[
+                    t * self._p : (t + 1) * self._p,
+                    col_start : col_end
+                ] = -1
+
+            # apply the given prior knowledge
+            for t in range(self._T):
+                for tau in range(n_taus):
+                    if t < tau:
+                        continue
+
+                    ix = np.ix_(
+                        np.arange(
+                            t * self._p,
+                            (t + 1) * self._p
+                        ),
+                        np.arange(
+                            (t - tau) * self._p,
+                            (t - tau + 1) * self._p
+                        )
+                    )
+
+                    temp = pk[ix]
+                    temp[self._Aknw[t, tau] == 0] = 0
+                    temp[self._Aknw[t, tau] == 1] = 1
+                    pk[ix] = temp
+
+            model = DirectLiNGAM(
+                prior_knowledge=pk,
+                measure=self._measure,
+                random_state=self._random_state
+            )
+            model.fit(X_t.T)
+
+            # split the estimated adjacency matrix
+            adj = np.array(np.split(model.adjacency_matrix_, self._T, axis=1))
+            adj = np.array(np.split(adj, self._T, axis=1))
+
+            # construct output matrices
+            adjs = np.zeros((self._T, n_taus, self._p, self._p))
+            for t in range(self._n_lags, self._T):
+                for lag in range(n_taus):
+                    adjs[t, lag] = adj[t, t - lag]
+            adjs[:self._n_lags] = np.nan
+            adjs[:, 1:] = adjs[:, 1:][:, ::-1]
+
+            # make causal_orders
+            causal_orders = []
+            for t in range(self._T):
+                if t < self._n_lags:
+                    causal_orders.append([np.nan for _ in range(self._p)])
+                    continue
+
+                # extract causal_order at time t
+                targets = range(t * self._p, (t + 1) * self._p)
+                filter_ = list(map(lambda x: x in targets, model.causal_order_))
+                causal_order = np.array(model.causal_order_)[filter_]
+
+                # make numbers start from zero
+                causal_order = causal_order - min(causal_order)
+                causal_orders.append(causal_order.tolist())
+
+            self._adjacency_matrices = adjs
+            self._causal_orders = causal_orders
 
-        # output B(t,t), B(t,t-τ)
-        self._adjacency_matrices = np.empty(
-            (self._T, 1 + self._n_lags, self._p, self._p)
-        )
-        self._adjacency_matrices[:, :] = np.nan
-        for t in range(self._n_lags, self._T):
-            self._adjacency_matrices[t, 0] = B_t[t]
-            for l in range(self._n_lags):
-                if t - l != 0:
-                    self._adjacency_matrices[t, l + 1] = B_tau[t, l]
-
-        self._residuals = np.zeros((self._T, self._n, self._p))
-        for t in range(self._T):
-            self._residuals[t] = N_t[t].T
-        self._causal_orders = causal_orders
         return self
 
     def bootstrap(self, X_list, n_sampling, start_from_t=1):
diff --git a/tests/test_longitudinal_lingam.py b/tests/test_longitudinal_lingam.py
index dc4339a..be16cfe 100644
--- a/tests/test_longitudinal_lingam.py
+++ b/tests/test_longitudinal_lingam.py
@@ -70,6 +70,10 @@ def test_fit_success():
     p_values = model.get_error_independence_p_values()
     resid = model.residuals_
 
+    # prior knowledge
+    pk = np.ones((3, 2, 4, 4)) * -1
+    model = LongitudinalLiNGAM(prior_knowledge=pk)
+    model.fit(X_list)
 
 def test_fit_invalid_data():
     # Different features
@@ -240,6 +244,17 @@ def test_fit_invalid_data():
     else:
         raise AssertionError
 
+    # prior knowledge
+    pk = np.ones((3, 2, 4, 4)) * -1
+    try:
+        # pk.shape[1] != n_lags + 1
+        model = LongitudinalLiNGAM(n_lags=10, prior_knowledge=pk)
+        model.fit(X_list)
+    except ValueError:
+        pass
+    else:
+        raise AssertionError
+
 def test_bootstrap_success():
     # causal direction: x0 --> x1 --> x3
     x0 = np.random.uniform(size=1000)