Serviços às empresas > Análise estatística
I have a sales analysis problem that will require statistical and mathematical expertise. The task is to calculate the best possible formula to determine the value of a group of products, based on the sales that were achieved from that group of products. The products (about 2000 in total) were all listed at the same price, and about 10% of the products were sold. The products are defined by two variables (a and b). My first assumption is that we can determine the quality of these products by applying some kind of formula to these two variables. For example: (a x 3.45) + (b) = Value. I have one year of sales data for these products. My second assumption is, that if we have the " correct " formula for the value for these products, the sales should decline (or increase) in line with the value of the domains. So, the best possible formula, will be the formula that shows the closest correlation in value and sales. I have the data in excel, like this: Products for sale: Productandnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; Aandnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; BProduct1andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; 390andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; 5.6etc. Products sold: Productandnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; Aandnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; BProduct456andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; 990andnbsp; andnbsp; andnbsp; andnbsp; andnbsp; 2.4etc. In the end, when we apply the formula, we should be able to do two histograms for Products for Sale and Products Sold, and see that the sales declined as the value decreased (and vice versa). I am expecting a Power Law distribution. I realize that there cannot be a perfect formula, but I know there can be a formula that gets " close ", because I tried a few formulas just based on intuition, and doing the histograms, I see a correlation between value and sales. I do not have enough statistical knowledge to solve this problem, but my guess is that there is probably something pretty standard to do this, for example some form of regression analysis.
Postado ao 5 de Fevereiro as 16h18
Collect data for all stocks that belong in the BSE 100 index and Nifty 50 index and provide in excel. The data required are the following: Symbol, Company Name, Industry, Price, Market Cap, Avg daily Vol (last 10 days), EPS (last quarter), EPS (same quarter last year), P/E, P/S, P/B, Dividend Yield, Revenue (last qtr), Revenue (same qtr last year)
Postado ao 4 de Fevereiro as 20h29