570 discussion post

Classmate #2 post.

Tori Rosengarten - Regression Forecasting

Contains unread posts

Victoria Rosengarten posted Nov 6, 2017 8:07 PM

Subscribe

There are many different professions and lines of business that could benefit from a regression based forecast in order to predict numbers. One in particular may be a proficient mortgage banker. If they collect a few years of data of the home loans they originate, they would notice a trend where people buy homes during certain seasons, and refinance more at certain interest rate. The mortgage banker would want to be able to predict the number of loans they could expect at a certain period of time. This would be important to help the banker predict the amount of hours they will be expected to work, but more so, this is important information as they are paid commission, so knowing what volume of loan will help them predict their income.

The independent variable that would be tracked would be the amount of the loan requested. They would likely find this information most useful to forecast their personal earnings as they are paid commission based on the loan amount, making this the most important factor. Other important factors would be whether the transaction was a refinance or a purchase, what season of the year the transaction took place, and the prevailing market interest rate at the time of the transaction. All of these dependent variables indicate a customer’s willingness to enter in to a loan and help shed light on what the most common transaction types are, and when.  This information would be collected based on the mortgage banker’s personal sales experience so the results would be specifically tailored to the banker’s skillset. It could be tested a season in advance after the information was collected. By forecasting the upcoming season, after the three months had passed, the mortgage banker could compare the predicted volume of loans to the actual amount of loans that were originated in said season. If the income is in a fair margin of error, this would become an important tool in forecast and predicting what the commission would be for the mortgage banker.

Based on level of success, this could potentially outspread to other mortgage bankers or other roles involved as well. Processors, who also work on a commission of loan amounts, would find just as much use in being able to forecast work and commission. In the same manner, other mortgage bankers on the floor would find use in forecasting what their sales would be in upcoming seasons.