Python代写 | Predict Model homework 1

本次美国代写是Python预测模型的一个homework

1. In Class we developed the graph below. Even if small, is there a noticable pattern that we can conclude by looking at this graph? If there is, explain the reason and logic behind such pattern.

2. Working with the housing dataframe, add the following four dummy variables to the set of regressors in the dataframe corresponding to the proximity to the ocean of the district:'<1H OCEAN’,’INLAND’ ‘ISLAND’, ‘NEAR BAY’. This will be the dataframe and model “with dummies”. Explain why can’t we add five dummy variables in this scenario.

3. Through cross-variables with ten folds, explain which model you thind is better to predict median house value. The model with dummies or the original model without dummies. Please explain any intuition behind your answer.

4. In the housing dataframe we added these three variables: rooms_per_household, bedroom_per_room, population_per_household. Through cross-validation with ten folds explain which model (a model with these three variavles and a model without them) you think is better to predict median house value, or if they are in general not too different from each other. Please explain any intuition you behind your answer.