Question

ADS uses its "big data" databases to supply data used to measure and predict consumer behavior. Assume that you are a business analyst working for Mega Attractions in Orlando, FL. Suggest a model that might be used to measure and predict guest spending behavior. Discuss how data would be collected and what the possible independent and dependent variables might be. Attach an Excel spreadsheet demonstrating the model.

Answer 1

Buying Decision-making is a psychological construct. This means that decisions of the consumers can be infered from their past buying pattern. Consumers are influenced by emotional and nonrational considerations which are very difficult to evaluate.Hence we look at their past buying patterns and will try to predict whether he will come next for the purchase or not. As a business analyst i would be initially make a classification of various products available eg: if its a business on selling tourist packages we will have a group of 30 packages to different destinations all around the world. Classify them as budget packages, little expensive, more expensive and luxury packages and put all the 30 in suitable ones..You need to how many times a customer had business with you. What type of products the customer preffered in the past. His income level and interest. Affordability which has correlation with interest in spending. You should defenitely consider the competitors offer in the particular area. you concentrate to do business. Segment the customers in your region based on their interest and income, religion, social group etc; Demographic profiling must be necessrilly done. expenditure on other items like repairs, education etc.

One technique that you can use is market basket analysis (is a modelling technique based upon the theory that if you buy a certain group of items, you are more (or less) likely to buy another group of items) .You can also perform the predictive analytics (analyze current data to make predictions about future). for the next best offer or product. Also a gap analysis to see whether the analytics you implemented is meeting the business requirments. The most suitable model for predicting the consumer behaviour is performing a logistic regression where the dependant variable is the dichotomous (takes onlly two values). in this case whether a customer will come again for the purchase or not. This model gives the probability of a customer coming again based on a set of independent variables like his income, preferences, basket of materials available etc.

refer NSSO or bureu of labor statistics for the data..