Sales Price Prediction Dashboard

    

Multiple Linear Regression is used to find the relationship between  quantitative variables.i.e dependent variables and an independent variable in order to predict the values of target variables .

For Linear regression model we have considered continuous data as independent variables and sales price which is the target variable as the dependent variable .The model is then trained on the train data set and the created model is assessed on the test results.This trained model is utilized to forecast the sales price.This predicted sales price and the actual sales price is plotted against each other on a scatter plot to analyze the test results .The performance is then evaluated using linear regression metrics like accuracy and root mean square error and the multiple linear equation.Once the model is ready we use the model to forecast the sales price by inputting our custom parameters .

Below is the dashboard built in Tableau showcasing the predictive analytics using linear regression .

The above dashboard is an amalgamation of Tableau and Python called Tabpy .

TabPy (the Tableau Python Server) is an Analytics Extension implementation which expands Tableau’s capabilities by allowing users to execute python scripts and saved functions via Tableau table calculations .

Features Of Sale Price Prediction Dashboard :

Forecasting :

With the help of Linear Regression , advanced analytics like forecasting of the sales price is computed on the basis of continuous independent variables.

Feature Importance:

 The most impacting independent variable for sales price is calculated using coefficient of correlation .The variable with highest coefficient is considered as the top variable affecting the sales price .

Predictive vs Actual Sales Price Analysis

After model training , the resultant model is testified on the test data to produce actual sales price and the predicted sales price on scatter plot .Hence we can compare the value of predicted price with our actual sales price and drill down the model accuracy for each value.