Homework answers / question archive / Problem To ensure there is no discrimination between employees, it is imperative for the Human Resources department of Delta Ltd

Problem To ensure there is no discrimination between employees, it is imperative for the Human Resources department of Delta Ltd

Statistics

Share With

Problem

To ensure there is no discrimination between employees, it is imperative for the Human Resources department of Delta Ltd. to maintain a salary range for each employee with similar profiles Apart from the existing salary, there is a considerable number of factors regarding an employee’s experience and other abilities to which they get evaluated in interviews. Given the data related to individuals who applied in Delta Ltd, models can be built that can automatically determine salary which should be offered if the prospective candidate is selected in the company. This model seeks to minimize human judgment with regard to salary to be offered. Goal & Objective: The objective of this exercise is to build a model, using historical data that will determine an employee's salary to be offered, such that manual judgments on selection are minimized. It is intended to have a robust approach and eliminate any discrimination in salary among similar employee profiles 

Pls note the inputs

(1)Evaluated

Criteria Ratings Points

1. Model building and interpretation.

a. Build various models (You can choose to build models for either or all of descriptive, predictive or prescriptive purposes) b. Test your predictive model against the test set using various appropriate performance metrics c.Interpretation of the model(s)

Need to mention the rationality behind selection of multilinear regression model. Model building approach (Multicollinearity – VIF, significant features & their selection process based on p-values & coefficient estimates etc.) should have been documented for linear model. In performance matrix, Adjusted R2 should have been computed as R2 may have noise in it as it starts increasing. R2 & Adj R2 should be closer for better model. Likewise, MAPE should be considered as it is better than RMSE, coz it’s the square of absolute values and robust to outliers. Various other models such as Lasso, Ridge regressions etc. should have tried in linear models as these models are robust to outliers and in case, we find high multicollinearity in the features. These are used when there is high multicollinearity among features. It adds a small squared biased factor to the variables, but reduces the variances.

2. Model Tuning

a.Ensemble modelling, wherever applicable b. Any other model tuning measures(if applicable) c. Interpretation of the most optimum model and its implication on the business

Efforts to improve the models (model tuning) are lacking. Should have tried many iterations in multilinear regression. Non-linear and Ensembling models such as RF regressor, ANN & XGBoost etc. should have tried. Only mentioned the significant featuress as per the model, should have given recommendations a per those.