Here is an attempt to understand the probability of discounted price of a stock by comparing it to historical price and fundamental correlation. Have made use of some of the new features of pine in developing this script (Such as matrix and new features of tables such as cell merge and tooltip).
Script makes use of the library written on matrix matrix
🎲 Process
Probability is measured in two angles 🎯 Absolute : Measure the percentile of price and fundamentals with respect to all time high. The difference between the two is measure of probability of stock being undervalued. 🎯 Drawdown : Measure the percentile of distance from all time high for both price and fundamentals. The difference between the two is used for depicting the probability of stock being undervalued.
🎲 Components
In short, the definitions of stats presented are as below
Even though the financial ratios of TSLA are not in par with most of the fundamentally strong stocks, it is indeed growing steadily and at its near all time high.
Here the base columns regarding fundamentals are mostly red. This means, company has suffered setback with respect to their financials and the company is not where it used to be. But, if you see the differential probabilities, it says 92% of being undervalued?
Well, this is due to the fact that NKLA's fundamentals suffered most of the time and they are always below par when compared to price. Hence, such kind of cases may interpret the stocks as undervalued. Hence, even if the probability of being undervalued is more, it does not guarantee the quality of the stock. We need to be mindful overall financials of the company and how they fare with general standards.
Moving forward
To understand value of trending stock, use Absolute Probability (marked with P). Ex. GOOG, MSFT, BRK.B etc.
To understand value of stock which has been recently suffered huge price drop, look at drawdown based probability (marked with D). Ex. BABA, FB, PYPL, SQ, ROKU etc.
Changes to getPercentileDrawdown function using yet another new pine method of binary search. Functionally, this does not change anything. But, improves the code performance. Thanks to lejmer for suggesting this improvement :)