As far in the project progress, I have done with the Breusch-Pagan test and sorted that there is no significant evidence of heteroskedasticity in the data analysis. In today’s blog, I will talk about the Cross-Validation, the Training error, and Test error.

An essential tool in data analysis and machine learning, cross-validation is used to assess the effectiveness of predictive models, gauge how effectively a model generalizes to new data, and address problems like overfitting. Cross-validation’s main principle is to divide your dataset into numerous subsets (folds), train and test the model using various combinations of these subsets, and then combine the results to provide a more accurate assessment of model performance.

K-Fold Cross-Validation, which divides the dataset into K nearly equal-sized folds (or partitions), is the most used type of cross-validation. Each fold is utilised as the test set once and the remaining K-1 folds are used as the training data as the model is trained and evaluated K times.

Important concepts like training error and test error are utilised to assess how well a predictive model is working. They are used to judge a model’s quality and generalisability, as well as how well it is performing.

I am going to perform these techniques and implement in my analysis for the 354 CDC data points and will post the results in the next blog.