Question

Data Science, I will give thumb up, thank you

What is one critical drawback to the MLR (multiple linear regression model) model (or any MLR model) for predicting shardnado hazard? What are some modifications that could improve on this issue?

sharknado hazard: the hazard of a sharknado, where 1 is very unlikely and 100 is highly likely

Answer 1

The disadvantage of using a MLR (Multiple Regression Model) usually comes down to the data being used. Two examples of this are using incomplete data and falsely concluding that a correlation is a causation.

Some of the drawbacks of this process are: -

• It is assumed that the cause and effect relationship between the variables remains unchanged but this assumption may not always hold good and hence estimation of the values of a variable made on the basis of the regression equation may lead to erroneous and misleading results.
• The functional relationship that is established between any two or more variables on the basis of some limited data may not hold good if more and more data are taken into consideration.
• The process involves very lengthy and complicated procedure of calculations and analysis.
• The process cannot be used in case of qualitative phenomenon viz. honesty, crime etc.

In the case of Sharknado Hazard, one of the drawback is that the model might not hold true because something to be likely or unlikely is not time dependent and hence in this regard a lot of variables need to be considered in order to get a proper correlation. The number of variables that will be sufficient can be predicted from the difference of the R-squared and the adjusted R-squared value. If the difference between the two is more than 10%, then the researcher needs to either add or remove some of the dependent variables in order to get a proper model. Hence, it becomes complicated to come to a final decision for how much variables we need to consider as an independent one in order to get a proper correlation in the data in order to proceed with the analysis.