Hello Prabhu, I dont comment on this forum much as I generally dont have the time. I saw your topic and thought it worth a response because it it quite a tricky one to get right (there are many failed attempts out there in the world). I have done the different statistical forcasting methods you are asking for in Excel. However this is not something that you can simply 'do'. For these to work you have to really REALLY understand both the data and the statistical method being used. The mathematics of each are not difficult and can be implemented with simple +-*/ SUM AVG, but has to be done uniquey to each data set. What works for one set of data in Excel will not work for another. The data has to be 'massaged' into the 'shape' you need it so that calculations can be done.
My suggestion is that you need to study the statistics and how they work then you will be able to apply the method to any data set quite easily. If you dont understand the method of each statistic you will end up with numbers that dont mean anything and you cant rely on them. Unfortunately tools that perform these calcuations fail to tell you how to understand your data in a way that suits the statistic being used, and even experienced statisticians can go very very wrong here. I know, I worked with a team of 20 PhD's on research projects (all in stats) and even they can get the tools wrong. None of them had trouble building it by hand FWIW and in the end it was more reliable to do it this way. So, where do you go from here? Step 1: choose one of the statistics (I would suggest Holts Winters or double exponential smothing). Build one by hand and check the math each step of the way through. Do this with a good statistics book and a known (well understood) set of data. Once you can do these step by step you can then apply that to any data you may have. Its a method not a 'formula' and so this approach will teach you also how to get your data ready to be analysed. I know this is probably not what you were looking for but I would suggest that anyone offering a 'simple' answer to this does not understand the topic at all. When you are doing forcasting you must approach the forcast very carefully and structure your data and thinking around the specific solution. Once you have solved a particular problem (created a model for a particular scenario) you can just keep feeding new data into it and the statistics you are working with with keep adapting to any changes - they are great this way. Take the time and you wont go wrong. Cheers and Good Luck The Frog -- Are you =EXP(E:RT) or =NOT(EXP(E:RT)) in Excel? And do you wanna be? It’s =TIME(2,DO:IT,N:OW) ! Join official Facebook page of this forum @ https://www.facebook.com/discussexcel FORUM RULES 1) Use concise, accurate thread titles. Poor thread titles, like Please Help, Urgent, Need Help, Formula Problem, Code Problem, and Need Advice will not get quick attention or may not be answered. 2) Don't post a question in the thread of another member. 3) Don't post questions regarding breaking or bypassing any security measure. 4) Acknowledge the responses you receive, good or bad. 5) Jobs posting is not allowed. 6) Sharing copyrighted material and their links is not allowed. NOTE : Don't ever post confidential data in a workbook. Forum owners and members are not responsible for any loss. --- You received this message because you are subscribed to the Google Groups "MS EXCEL AND VBA MACROS" group. To unsubscribe from this group and stop receiving emails from it, send an email to excel-macros+unsubscr...@googlegroups.com. To post to this group, send email to excel-macros@googlegroups.com. Visit this group at http://groups.google.com/group/excel-macros. For more options, visit https://groups.google.com/groups/opt_out.
