# Chosing a model to forecast free cash flow of a company

#### pashtun

##### New Member
Dear statiticians, experienced users,

Recently I have started trading in the stock market. An important element in choosing a stock is the intrinsic value, atleast for me. To calculate this intrinsic value, one needs to forecast the future cash flow based on past performance. I thought this would be a nice experiment to try with modeling.

I have little experience with this subject, a long time ago I used latent class mixed modeling (lcmm) to find subclasses in a population. I thought i could use this function again, as it gives you a mean condidance interval of the predicted values with a certain random effect. I would like to repeat this using only "1" subclass (therefore none actually), however the model uses data tied to certain "ID", since I am using this to calculate the same company and not different individuals like last time, I am doubting wether this is the best approach.

Therefore my question is: what model would you use to predict the future cash flow of a company if you wanted to add a certain random effect? GLMM?

There is more than one way to calculate free cash flow. You can predict free cash flow directly, for facebook this would be:

Time 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
FCF 405 943 377 2,860 3626 6,076 11,617 17,483 15,359 21,212 18,823

Numbers are in millions of dollars. Ideally you want to predict 5 years in to the future or 10 if you are less conservatie in your estimations.

Another way to calculate free cash flow, is to predict the revenue (which is easier) and then use the 'average revenue margin' to calculate free cash flow

Revenu 1974 3771 5089 7872 12466 17928 27638 40653 55383 70697 75157
Revenue/FCF 0.21 0.25 0.07 0.36 0.29 0.34 0.42 0.43 0.28 0.30 0.25

For example, if one predicted that the free cash flow was 80000 in 2021 and chose to use a conservative revenue margin of say 0.25, the predicted free cash flow in 2012 would be 20000 (80000*0.25).

Any idea's would be great and much appreciated. Since there may come as second corona wave and I have no way of predicting how it would impact the economy further, I would like to use a random effect.

Greetings

Pashtun