In real estates, housing market prediction (forecasting) is crucial

Subject:BusinessPrice:18.89 Bought3

Share With

In real estates, housing market prediction (forecasting) is crucial. There are many factors that may influence the house prices. The datasets housing.training.csv and housing.testing.csv contain 25 quantitative explanatory variables describing many aspects of residential homes in Ames, IA.

The goal of this project is to predict house prices. To this end, we will be using regression analysis.

1. In Week 4 Portfolio Milestone, you've examined housing.training.csv dataset. Now, examine housing.testing.csv dataset and perform the same tasks as given in Week 4 Portfolio Milestone. Using R, calculate the summary statistics (minimum, maximum, mean, median, and standard deviation) and create a histogram of sale price for each dataset. Comparing with housing.training,csv dataset, describe the similarities and/or differences.
2. Combine the two datasets housing.training.csv and housing.testing.csv. This can be done in R by using the function combine(). Create a histogram of sale prices for the combined dataset and compare it with the histograms from training and testing datasets. Describe the similarities and differences.
3. Using only the dataset housing.training.csv, fit a linear regression model using all the explanatory variables and SalePrice as the response variable.
4. What are the significant factors? How do these variables relate to the sale price? Interpret your estimated model.
5. Remove all the rows with missing values (NA) from the dataset housing.testing.csv. The function complete.cases() can be used. Using only the first 20 rows from housing.testing.csv, predict the sale price. The R function predict() can perform this task. You should have 20 predicted sale prices.
6. Compare the predicted sale prices to the actual sale prices from the housing.testing.csv dataset (the first 20 rows). How good is your prediction?

For each R output result, you may either type directly into a Word document or take a screenshot. If you take the screenshot, make sure that the current date is shown.

Ensure everything is clearly labeled. The report must be 10-12 pages long, including a title page and reference page (the report itself should be 8-10 pages). Cite 2-3 academic sources other than the textbook, course materials, or other information provided as part of the course materials. Follow APA format, according to CSU Global Writing Center (Links to an external site.).

1 Module 4: Option #1: Logistic Regression Module 4: Option #1: Logistic Regression Carlos Figueroa Colorado State University Global MIS470: Data Science Foundation Kelly Wibbenmeyer 4-11-2021 Module 4: Option #1: Logistic Regression 2 Module 4: Option #1: Logistic Regression After creating the scatter plot of automatic or manual transmission versus mpg we can see that the data is either a one or a zero. From this data we try to see the correlation between the type of transmission the car has versus how much gas it saves or spends but we can only tell from the scatter plot that automatic seems to get better mpg or skews towards higher mpg. This is one of the reasons that a simple linear regression model may not fit our analysis because it is not always representing a complete description of the relationship amongst the variables (Flom, 2019). There are many factors this graph does not show that could attribute to the variance in mpg such as the mt car being a sports car, having a bigger engine, or being a bigger car in general just to name a few factors. Module 4: Option #1: Logistic Regression 3 We would classify the transmission as automatic when we test with the mpg value of 16 because of how low the probability of it being a manual is. With their only being 2 variables of 1 or 0 for Module 4: Option #1: Logistic Regression the transmission it forces us to use 0.5 as the cut off value which the car with 16 mpg likely exceeds. 4 Module 4: Option #1: Logistic Regression 5 References Flom, P. (2019, March 2). The Disadvantages of Linear Regression. Sciencing. https://sciencing.com/disadvantages-linear-regression-8562780.html. 1 Module 4: Option #1: Logistic Regression Module 4: Option #1: Logistic Regression Carlos Figueroa Colorado State University Global MIS470: Data Science Foundation Kelly Wibbenmeyer 4-11-2021 Module 4: Option #1: Logistic Regression 2 Module 4: Option #1: Logistic Regression After creating the scatter plot of automatic or manual transmission versus mpg we can see that the data is either a one or a zero. From this data we try to see the correlation between the type of transmission the car has versus how much gas it saves or spends but we can only tell from the scatter plot that automatic seems to get better mpg or skews towards higher mpg. This is one of the reasons that a simple linear regression model may not fit our analysis because it is not always representing a complete description of the relationship amongst the variables (Flom, 2019). There are many factors this graph does not show that could attribute to the variance in mpg such as the mt car being a sports car, having a bigger engine, or being a bigger car in general just to name a few factors. Module 4: Option #1: Logistic Regression 3 We would classify the transmission as automatic when we test with the mpg value of 16 because of how low the probability of it being a manual is. With their only being 2 variables of 1 or 0 for Module 4: Option #1: Logistic Regression the transmission it forces us to use 0.5 as the cut off value which the car with 16 mpg likely exceeds. 4 Module 4: Option #1: Logistic Regression 5 References Flom, P. (2019, March 2). The Disadvantages of Linear Regression. Sciencing. https://sciencing.com/disadvantages-linear-regression-8562780.html. 1 Module 4: Portfolio Milestone: Option 1 Module 4: Portfolio Milestone: Option 1 Carlos Figueroa Colorado State University Global MIS470: Data Science Foundation Kelly Wibbenmeyer 4-11-2021 Module 4: Portfolio Milestone: Option 1 Module 4: Portfolio Milestone: Option 1 2 Module 4: Portfolio Milestone: Option 1 3 The distribution of the SalePrice is right skewed so we see that most of the data for sales prices is centered around 150-200 thousand dollars. Most of the house were sold at a median price whole very few were sold any higher than 400,000 dollars.