Statistical Experiments for Data Scientists

Designing and interpreting experiments to understand customer behavior, sales, profitability, advertisement effectiveness etc. is part and parcel of data scientist’s job. Here I’ll illustrate with a simple example how statistical experiments can be used in a business context.

Observation:

Cab cancellation in city A is 150% higher in the month of November than average car cancellation over the year.

Hypothesis:

The higher cancellation is due to a seasonal factor affecting cancellations in November

Experiment:

Calculate average cancellation rates during the year and during the month of November

Hypothesis Testing

Based on the results of this experiment the Data Scientist can determine whether or not this difference is significant enough to warrant increased attention and focus from the management.

Leave a Reply

Your email address will not be published.