###

What is Random Forest Regression?

Random Forest Regression creates a set of Decision Trees from a randomly selected subset of the training set, and aggregates by averaging values from different decision trees to decide the final target value.

Random Forest Regression is limited to predicting numeric output so the dependent variable has to be numeric in nature. The minimum sample size is 20 cases per independent variable.

To further clarify the use of the Random Forest Regression model, let’s look at a sample analysis to optimize house pricing, based on numerous variables:

Explore the use cases below, to better understand the value of Random Forest Regression.

Business Use Case – House Price

Business Problem: A real-estate brokerage company wants to measure the impact of locality, the number of rooms, the area(sq. yards) etc. on a house price. The goal of this statistical analysis is to help us understand the relationship between house features and how these variables are used to predict house price.

Target: House Price

Predictors: Area with carpet, Rainfall, city, parking, distance from hospital, distance from shopping, etc.

• The business can determine which predictors have a significant impact on house price.
• Pricing strategies and recommendations will be more accurate and result in quicker sales.
• If the number of rooms or the distance from shopping or schools are significant factors, these factors are given more focus when searching for a house that fits a client budget and affects profit.

Business Problem: An agriculture business wants to measure the impact of weather, market price, quality of crop, land used etc. on the crop price.

Input Data: Predictor/Independent Variables

• Weather
• Demand
• Crop health

Dependent Variable: Crop Price

• Business can clarify which factors have a significant impact on crop price.
• Pricing strategies can be refined to improve accuracy and meet targeted crop pricing and revenue.
• If crop health and climate are significant factors, these factors would receive more focus when deciding crop price.

Business User Case – Compensation Policies

Business Problem : A business wishes to measure the salary of employee based on position, experience, degree, level, productive hours etc.

Input Data: Predictor/Independent Variables

• Position
• Years of experience
• Productive hours

Dependent Variable: Salary of Employee