Great idea to start an information and educational journey. I totally agree with your approach. In my investment journey I have followed online education by Aswath Damodaran from The Stern Business School NYU. I also have a keen interest in quantitative models, and I have produced a model that ranges companies by three parameters: Next year free cash flow to EV, NOPAT to EV, and Next year growth in revenues. I have completed a succesful back test on the model for US stocks 2011-2023. I will follow your post with great interest.
Yes I use my ranking method as a sort of screener for interesting stocks. My utilation of the methode is still fairly new, as I only had the model ready in mid november 2024. Thus I can not yet make a claim of a long track record for the model. I have plans for additional parameters in the model for instance adjusting for R&D and implementing growth in EBIT-margins.
Great post! Are you going to share how you model out free cash calculations, for us DIY investors? It seems to me that you're circling around the DCF model, which honestly seems old school as well, although it might still be the best way to value a company?
Persoanlly I have a PE / EPS approach, which I'm very happy to be challenged on ✌️
The DCF model can trace its origin all the way back to the 1930s, but it is more complicated than valuation multiples, which made it less attractive to do by hand leading to simpler heuristics like the P/E ratio etc. which were successfully popularized by Benjamin Graham etc.
But the fact that something is "old school" is not an argument for not applying the method. Linear algebra and especially matrix multiplication is around 180 years old techniques applied in every computational system including LLMs. The very fact that it is old and is still applied means that it will continue to be used for a long time (Lindy effect).
The DCF model is theoretically the right way to model, but it comes with uncertainty which is just a given when we try to value an asset with a very long life. But I will show later with an example how things are done in a DCF model and also why the "inversion method" to DCF is super important to limit any biases in the model.
The issue with P/E and EPS are these...
P/E: when have estimates of next year's EPS, so investors only need to apply an appropriate P/E, but since the we already have the E in P/E the only unknown variable is price P. In other words, you can set value to whatever you want and thus the P/E framework because a useless tautology. We need a framework to estimate value and you not getting that through P/E and EPS.
EPS: does not account for the cost of capital, exclude the incremental investments in working capital and fixed capital which are needed for growth, management can game EPS figures but not cash flows
All of these things will be clearer as I publish more posts :-)
Great idea to start an information and educational journey. I totally agree with your approach. In my investment journey I have followed online education by Aswath Damodaran from The Stern Business School NYU. I also have a keen interest in quantitative models, and I have produced a model that ranges companies by three parameters: Next year free cash flow to EV, NOPAT to EV, and Next year growth in revenues. I have completed a succesful back test on the model for US stocks 2011-2023. I will follow your post with great interest.
Cool Anders and thanks for your kind words. Are you investing based on your methodology?
Yes I use my ranking method as a sort of screener for interesting stocks. My utilation of the methode is still fairly new, as I only had the model ready in mid november 2024. Thus I can not yet make a claim of a long track record for the model. I have plans for additional parameters in the model for instance adjusting for R&D and implementing growth in EBIT-margins.
Great post! Are you going to share how you model out free cash calculations, for us DIY investors? It seems to me that you're circling around the DCF model, which honestly seems old school as well, although it might still be the best way to value a company?
Persoanlly I have a PE / EPS approach, which I'm very happy to be challenged on ✌️
The DCF model can trace its origin all the way back to the 1930s, but it is more complicated than valuation multiples, which made it less attractive to do by hand leading to simpler heuristics like the P/E ratio etc. which were successfully popularized by Benjamin Graham etc.
But the fact that something is "old school" is not an argument for not applying the method. Linear algebra and especially matrix multiplication is around 180 years old techniques applied in every computational system including LLMs. The very fact that it is old and is still applied means that it will continue to be used for a long time (Lindy effect).
The DCF model is theoretically the right way to model, but it comes with uncertainty which is just a given when we try to value an asset with a very long life. But I will show later with an example how things are done in a DCF model and also why the "inversion method" to DCF is super important to limit any biases in the model.
The issue with P/E and EPS are these...
P/E: when have estimates of next year's EPS, so investors only need to apply an appropriate P/E, but since the we already have the E in P/E the only unknown variable is price P. In other words, you can set value to whatever you want and thus the P/E framework because a useless tautology. We need a framework to estimate value and you not getting that through P/E and EPS.
EPS: does not account for the cost of capital, exclude the incremental investments in working capital and fixed capital which are needed for growth, management can game EPS figures but not cash flows
All of these things will be clearer as I publish more posts :-)
That's a very good primer for all of us PG! Thank you!
Thanks Theo!