Follow the instructions posted at https://ds202-at-isu.github.io/labs.html for the lab assignment. The work is meant to be finished during the lab time, but you have time until Monday evening to polish things.
Include your answers in this document (Rmd file). Make sure that it knits properly (into the md file). Upload both the Rmd and the md file to your repository.
All submissions to the github repo will be automatically uploaded for grading once the due date is passed. Submit a link to your repository on Canvas (only one submission per team) to signal to the instructors that you are done with your submission.
<<<<<<< HEAD
| TL;DR Summary - Bhargav Yellepeddi: |
| The analysis of Ames property sales since 2017 shows most sales between $100K–$400K. Total living area follows a normal distribution, with outliers. A positive correlation exists between living area and sale price, though it weakens for larger properties. Missing data led to 447 rows being excluded. |
| ======= |
TL;DR Summary - Bhargav Yellepeddi:
The analysis of Ames property sales since 2017 shows most sales between $100K–$400K. Total living area follows a normal distribution, with outliers. A positive correlation exists between living area and sale price, though it weakens for larger properties. Missing data led to 447 rows being excluded.
TL;DR Summary - Keenan Jacobs:
In this analysis, we explore the relationship between property size (Acres) and sale price (Sale Price). We find that larger properties tend to have higher sale prices on average. However, there are some properties with large acreage that have surprisingly low sale prices, which could be due to factors such as location, property condition, or zoning regulations.
Keenan Jacobs - SalePrice and Acres
ca7ece837b2724149161e68e0bc46b542e384842
Bhargav Yellepeddi - SalePrice and Total Living Area Analysis:
# Load the Ames dataset
data <- classdata::ames# View first few rows
head(data)## # A tibble: 6 × 16
## `Parcel ID` Address Style Occupancy `Sale Date` `Sale Price` `Multi Sale` YearBuilt Acres `TotalLivingArea (sf)` Bedrooms FinishedBsmtArea (sf…¹ `LotArea(sf)` AC FirePlace Neighborhood
## <chr> <chr> <fct> <fct> <date> <dbl> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <chr> <chr> <fct>
## 1 0903202160 1024 RIDGEWOOD AVE, AMES 1 1/2 Story … Single-F… 2022-08-12 181900 <NA> 1940 0.109 1030 2 NA 4740 Yes Yes (28) Res: B…
## 2 0907428215 4503 TWAIN CIR UNIT 105, AMES 1 Story Frame Condomin… 2022-08-04 127100 <NA> 2006 0.027 771 1 NA 1181 Yes No (55) Res: D…
## 3 0909428070 2030 MCCARTHY RD, AMES 1 Story Frame Single-F… 2022-08-15 0 <NA> 1951 0.321 1456 3 1261 14000 Yes No (32) Res: C…
## 4 0923203160 3404 EMERALD DR, AMES 1 Story Frame Townhouse 2022-08-09 245000 <NA> 1997 0.103 1289 4 890 4500 Yes No (31) Res: M…
## 5 0520440010 4507 EVEREST AVE, AMES <NA> <NA> 2022-08-03 449664 <NA> NA 0.287 NA NA NA 12493 No No (19) Res: N…
## 6 0907275030 4512 HEMINGWAY DR, AMES 2 Story Frame Single-F… 2022-08-16 368000 <NA> 1996 0.494 2223 4 NA 21533 Yes Yes (37) Res: C…
## # ℹ abbreviated name: ¹`FinishedBsmtArea (sf)`
# Inspect the structure of the dataset
str(data)## tibble [6,935 × 16] (S3: tbl_df/tbl/data.frame)
## $ Parcel ID : chr [1:6935] "0903202160" "0907428215" "0909428070" "0923203160" ...
## $ Address : chr [1:6935] "1024 RIDGEWOOD AVE, AMES" "4503 TWAIN CIR UNIT 105, AMES" "2030 MCCARTHY RD, AMES" "3404 EMERALD DR, AMES" ...
## $ Style : Factor w/ 12 levels "1 1/2 Story Brick",..: 2 5 5 5 NA 9 5 5 5 5 ...
## $ Occupancy : Factor w/ 5 levels "Condominium",..: 2 1 2 3 NA 2 2 1 2 2 ...
## $ Sale Date : Date[1:6935], format: "2022-08-12" "2022-08-04" "2022-08-15" "2022-08-09" ...
## $ Sale Price : num [1:6935] 181900 127100 0 245000 449664 ...
## $ Multi Sale : chr [1:6935] NA NA NA NA ...
## $ YearBuilt : num [1:6935] 1940 2006 1951 1997 NA ...
## $ Acres : num [1:6935] 0.109 0.027 0.321 0.103 0.287 0.494 0.172 0.023 0.285 0.172 ...
## $ TotalLivingArea (sf) : num [1:6935] 1030 771 1456 1289 NA ...
## $ Bedrooms : num [1:6935] 2 1 3 4 NA 4 5 1 3 4 ...
## $ FinishedBsmtArea (sf): num [1:6935] NA NA 1261 890 NA ...
## $ LotArea(sf) : num [1:6935] 4740 1181 14000 4500 12493 ...
## $ AC : chr [1:6935] "Yes" "Yes" "Yes" "Yes" ...
## $ FirePlace : chr [1:6935] "Yes" "No" "No" "No" ...
## $ Neighborhood : Factor w/ 42 levels "(0) None","(13) Apts: Campus",..: 15 40 19 18 6 24 14 40 13 23 ...
unique(data[,"Sale Price"])## # A tibble: 1,327 × 1
## `Sale Price`
## <dbl>
## 1 181900
## 2 127100
## 3 0
## 4 245000
## 5 449664
## 6 368000
## 7 110000
## 8 350000
## 9 242000
## 10 293000
## # ℹ 1,317 more rows
# Histogram of SalePrice
ggplot(data, aes(x = `Sale Price`)) +
geom_histogram(binwidth = 20000, fill = 'blue', color = 'black') +
labs(title = "Distribution of Sale Prices", x = "Sale Price", y = "Count")
# Summary statistics of TotalLivingArea
summary(data$`TotalLivingArea (sf)`)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0 1095 1460 1507 1792 6007 447
# Histogram of TotalLivingArea
ggplot(data, aes(x = `TotalLivingArea (sf)`)) +
geom_histogram(binwidth = 100, fill = 'green', color = 'black') +
labs(title = "Distribution of Ground Living Area", x = "Ground Living Area (sq ft)", y = "Count")## Warning: Removed 447 rows containing non-finite outside the scale range (`stat_bin()`).
# Scatterplot of SalePrice vs TotalLivingArea
ggplot(data, aes(x = `TotalLivingArea (sf)`, y = `Sale Price`)) +
geom_point(color = 'purple') +
labs(title = "Sale Price vs Ground Living Area", x = "Ground Living Area (sq ft)", y = "Sale Price")<<<<<<< HEAD
## Warning: Removed 447 rows containing missing values or values outside the scale range (geom_point()).
Thien Tam Nguyen - SalePrice and LotArea Analysis:
1.
“Sale Price”: A numeric variable representing the selling price of each property. This variable ranges from low-value properties to luxury ones, generally from around $50,000 to over $20 million as seen in the histogram.
“Lot Area (sf)”: A numeric variable representing the size of the property lot in square feet. The Lot Area ranges from small plots (less than 10,000 square feet) to very large areas exceeding 500,000 square feet.
Both variables are continuous, and they provide insight into the property values and the physical size of the lots.
-
I am taking analysis of sale price and lotArea
-
The Sale Price histogram shows that the majority of properties are sold for prices under $5 million, with most sales clustered around lower price points.There are a few extremely high sale prices (over $15 million), which could indicate unique or luxury properties.The pattern shows that the market for properties in Ames is mostly concentrated in the lower price range, with a small number of high-value outliers.
-
The Sale Price has a wide range, but most properties are concentrated in the lower value range, with some significant outliers.The Lot Area is also skewed, with most properties being small to medium-sized, but a few very large properties exist.There is a general positive trend between Lot Area and Sale Price, but the relationship is not perfectly linear, and other factors such as location or house features may play a crucial role.
# Load Ames dataset
data <- classdata::ames# View first few rows for relevant columns with corrected names
colnames(data)## [1] "Parcel ID" "Address" "Style" "Occupancy" "Sale Date" "Sale Price" "Multi Sale" "YearBuilt"
## [9] "Acres" "TotalLivingArea (sf)" "Bedrooms" "FinishedBsmtArea (sf)" "LotArea(sf)" "AC" "FirePlace" "Neighborhood"
head(data[, c("Sale Price", "LotArea(sf)")])## # A tibble: 6 × 2
## `Sale Price` `LotArea(sf)`
## <dbl> <dbl>
## 1 181900 4740
## 2 127100 1181
## 3 0 14000
## 4 245000 4500
## 5 449664 12493
## 6 368000 21533
# Inspect structure to understand data types of specific columns
str(data[, c("Sale Price", "LotArea(sf)")])## tibble [6,935 × 2] (S3: tbl_df/tbl/data.frame)
## $ Sale Price : num [1:6935] 181900 127100 0 245000 449664 ...
## $ LotArea(sf): num [1:6935] 4740 1181 14000 4500 12493 ...
# Histogram of Lot Area
ggplot(data, aes(x = `LotArea(sf)`)) +
geom_histogram(binwidth = 1000, fill = 'orange', color = 'red') +
labs(title = "Distribution of Lot Area in Ames Properties",
x = "Lot Area (square feet)",
y = "Count")## Warning: Removed 89 rows containing non-finite outside the scale range (`stat_bin()`).
#Scatterplot of Sale Price vs Lot Area in Ames Properties
ggplot(data, aes(x = `LotArea(sf)`, y = `Sale Price`)) +
geom_point(color = 'darkgreen') +
labs(title = "Sale Price vs Lot Area in Ames Properties",
x = "Lot Area (square feet)",
y = "Sale Price ($)")## Warning: Removed 89 rows containing missing values or values outside the scale range (`geom_point()`).
The scatter plot of Sale Price vs Lot Area shows a positive relationship, indicating that larger lot sizes generally lead to higher sale prices. However, there are many points with large lot areas and lower-than-expected sale prices, implying that other factors may be influencing the value.
# Scatterplot with trend line
ggplot(data, aes(x = `LotArea(sf)`, y = `Sale Price`)) +
geom_point(color = 'darkblue') +
geom_smooth(method = "lm", color = 'red', se = FALSE) +
labs(title = "Sale Price vs Lot Area with Trend Line",
x = "Lot Area (square feet)",
y = "Sale Price ($)")## `geom_smooth()` using formula = 'y ~ x'
## Warning: Removed 89 rows containing non-finite outside the scale range (`stat_smooth()`).
## Warning: Removed 89 rows containing missing values or values outside the scale range (`geom_point()`).
The scatter plot with the trend line shows a downward slope, which could indicate that there isn’t a straightforward, consistent linear relationship between Lot Area and Sale Price for all properties. This suggests variability and possibly the influence of other characteristics.
# Histogram of Sale Price
ggplot(data, aes(x = `Sale Price`)) +
geom_histogram(binwidth = 25000, fill = 'skyblue', color = 'black') +
labs(title = "Distribution of Sale Prices in Ames Properties",
x = "Sale Price ($)",
y = "Count")
=======
## Warning: Removed 447 rows containing missing values or values outside the scale range
## (geom_point()).
Keenan Jacobs and Analysis of Acres
summary(data$Acres)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 0.0000 0.1502 0.2200 0.2631 0.2770 12.0120 89
The range shows a semi-wide distribution, with most properties having a small land size. Let’s visualize this distribution using a histogram.
# Histogram of Acres
ggplot(data, aes(x = Acres)) +
geom_histogram(binwidth = 0.1, fill = 'steelblue', color = 'black') +
labs(title = 'Distribution of Property Size (Acres)', x = 'Acres', y = 'Frequency')## Warning: Removed 89 rows containing non-finite outside the scale range
## (`stat_bin()`).
# Scatterplot of Acres vs Sale Price
ggplot(data, aes(x = Acres, y = `Sale Price`)) +
geom_point(alpha = 0.5, color = 'darkgreen') +
labs(title = 'Scatterplot of Property Size vs Sale Price', x = 'Acres', y = 'Sale Price ($)')## Warning: Removed 89 rows containing missing values or values outside the scale range
## (`geom_point()`).
Here we can see there is a few huge outliers so we should probably filter them out and look again out our data as well as look at our outliers.
# Filter properties with high Acres but low Sale Price
anomalies <- data %>% filter(Acres > 1, `Sale Price` < 50000)
filtered_data <- data %>%
filter((Acres < 5 & `Sale Price` < 500000))# Boxplot of Anomalies by Style
ggplot(anomalies, aes(x = Style, y = `Sale Price`)) +
geom_boxplot(fill = 'lightgreen', color = 'darkgreen') +
coord_flip() + # Flip coordinates for better readability
labs(title = 'Distribution of Sale Prices for Anomalies by Style',
x = 'Property Style',
y = 'Sale Price ($)') +
theme_minimal()
# Plotting the filtered data to confirm anomalies have been removed
ggplot(filtered_data, aes(x = Acres, y = `Sale Price`)) +
geom_point(alpha = 0.5, color = 'darkgreen') +
labs(title = 'Scatterplot of Property Size vs Sale Price (Filtered Data)',
x = 'Acres', y = 'Sale Price ($)')
Here we can see a few points of data looking at the anomolies and also looking at the cleaned data. From the filtered data, we cna see that there is little correlation between acre size and sales price.
ca7ece837b2724149161e68e0bc46b542e384842