Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
relevant fundamental factors?

Wondering if anyone would be willing to suggest a set of fundamental factors that might be germane to the Q contest/fund? Looking at https://www.quantopian.com/help/fundamentals, there is a dizzying number of fields, and I have to think many of them just don't matter.

An example is free cash flow yield, which one site describes as "Free Cash Flow Yield: The Best Fundamental Indicator." Are there others that one could test that are well-known in the investment world (and not one of the Q risk model factors)? I'm not looking for any "secret sauce" here--just some ideas of what is commonly known/believed in the investment world.

10 responses

Hey Grant,
Here's a short list of some fundamentals I've been working with trying to rank by each sector.

Alpha Generation & Compute Z-scores:

    # Working Capital to Assets  
    # Retained Earnings to Assets  
    # EBITA to Assets  
    # Sales to Assets  
    # Mkt Cap to Liabilities  

Thanks Daniel -

I'll add them to my list of things to investigate.

Hi Grant,

Basically, valuation ratios are most commonly used when 'pricing' securities.
- Present value models (Dividens / FCFE / FCFF), require to compute CAPM required return on equity and the growth rate
- Multipliers models (P/E, P/S, P/B...)
Other models to consider:
- Altman z-score (credit risk)
- Beinish M-score (reporting quality & account manipulations)
- ROE decomposition

Hi Daniel and Mathieu,

May I know how you guys come up with the list of those fundamentals? Is there any source materials like research or books you think that I should read to know more about them?

Hey,

Personally, I learned financial reporting/analysis during my studies, and with the CFA curriculum, but you can find the basics almost everywhere (by typing 'Fundamental valuation methods" on google for example).

On the Fundamental Data Reference page, there is an Asset Classification section which has some derived fundamental indicator scores/grades from Morningstar. I gather that this is all about the Morningstar Style Box. I'm wondering if the Q risk factors allow for any style-box tilt, or if the idea is to be style-box neutral?

I'm curious if fundamentals are all that useful in the first place. In theory, the financial companies that melted down in 2008 (e.g. Lehman) should have had terrible fundamentals over the preceding years. I'm guessing not a whole lot has changed since then, and so one could have spectacular fundamentals and the company could still go "poof" without much warning. It seems that companies are always going to be playing varying degrees of accounting razzmatazz to hide whatever they can from the market, for as long as they can, hoping that the day of reckoning never comes (at least on their watch). I don't have a business background, but the recent Great Recession would seem to be an existence proof that what matters may not be captured in company fundamentals.

Hi Grant,

You're confusing between detecting financial crisis / collapses which are linked to black-swan risk types (combined with asset value manipulations as we experienced during 2007-2008) and finding overvalued / undervalued or even financial distress in 'economic and financial stability' periods. These are two different situations. The first one is almost impossible to predict (by definition of black swan events as described by N.Taleb), however in the second, bankruptcies may occur because of fundamentals deterioration (even if there is many other reasons why a company may go bankrupt in practice), and in this case, the deterioration is somewhat predictable when studying the evolution of fundamentals or economic environment (take a look at what is happening to the french retail group Casino at the moment).
I would add that, on a macro view, global recessions are unpredictable, on a micro view however, companies' financial health, and thus over and under valuations, can be. Since fundamentals alone can't explain a company financial health, you also need to focus on other factors (like economic environment for example), this is why in practice, reduced-form models have more predictive power than structural or accounting models when assessing credit risk.

Hope it helps

Thanks, Grant and Mathieu, for your insights.

I also pulled up Lehman Brother annual report to see their fundamentals before the collapse.

Both their ROE and margin decreased during 2017 relative to last year. Their leverage ratio also went above 30x. Hence, during the same period, their stock price has already been deteriorating.

After they released the 10-Q in 2018, despite their leverage ratio being reduced to 24x (which they faked by tweaking their REPO), they still reported a negative revenue. Their stock further plummeted till they bankrupted in September.

So, even though they fraudulently tweaked some parts of their financial statement, other Fundamentals that are harder to manipulate such as revenue and fundamentals from previous periods still foreshadowed their increasingly bad performance.

Thanks all -

I suppose there is always some fraction of the current stock price that is explained by the fundamentals of the company. There is some price the market would pay for the fundamentals, assuming no other information were available. But there are plenty of other factors that the market will bring into play to sort out how much it is willing to pay. Another case within my lifetime was the dot-com/telecom bubble and its irrational exuberance, where stock prices became completely unhinged from fundamentals.

I read the Black Swan book awhile back. Ironic that it was published in 2007.

The whole system seems a bit unhinged. Take executive pay, for example. Here, it says:

In 2016, the CEOs of the top 350 U.S. firms earned on average $15.6 million.

Now what does anyone need $16 M per year for? One individual doesn't even need $16 M in a lifetime. Completely out of whack.