Readings for 9/13/18

For our class discussion Thursday we were assigned to read chapters 9, 10, and 11. Chapter 9 talks about predictive analytics and how businesses use them regularly . Chapter 10 talks all about regression analyst, how they are used, their benefits, and errors that commonly get made. Chapter 11 focuses on correlations and the risks behind correlations with business's.

One thing I found interesting in chapter 9 was the use behind predicative analytics and how they are used more often then I thought. The reading gives examples like Customer lifetime value, and forecasts of sales for the next quarter, and I never realized how important this is in every business. I made a personal connection with this aspect of chapter 9 because I'm a very habitual buyer, with products and the store I get it from. Because of this, I see how important I am as a customer from the various places I spend my money at. Whether it is gas for my car, deli sandwiches, or the clothing I wear, I am very important to the customer lifetime value of the stores I go to. One thing I didn't fully understand was the data aspect of the predictive analytics. The author says that its important to make sure you use good data, but how does one specifically understand what is good data or bad? Without the knowledge of knowing what is good data or bad, I feel like the risk would still always be there to analyze it.

Another thing I found interesting in chapter 10 was the example used to explain regression analyst. Not fully understanding how regression analyst are used or when they are supposed to be used, the example broke it down to help me learn. The example used a business manager  trying to predict next months sales. I now understand why regression models are so important for business's as the variables that relate to the cause almost usually show a correlation. In the example there was a clear correlation with sales and rain fall, which directly effects their sales and how they go about promotions. The question I have for chapter 10 is the reading says that you must be specific to the data analyzer, but in what ways can you be most specific to not run into problems?

Chapter 11 teaches the importance of a correlation and when to act on it or not. What I found most interesting was the two most important reasons to decide if a correlation is worth taking a chance on. Those reasons were that confidence that the correlation will reliably happen again in the future, and the trade off between how big the risk and reward is. I found these important because I didn't know that before, and it stuck with me when thinking about future business risks myself. One example I think of is real estate, where my best friend and moms boyfriend are both heavily involved in. Correlations within the market, what other houses were sold for, and what are other houses are available definitely have a huge role in deciding what properties people end up with next. The question I have for this chapter is one chart shows that its worth it to go through with correlations when the benefits outweigh the risks, but is that always the case? is their ever cases of correlation where everything can look perfect but acting may still not be the best idea?