On Investment Modeling, Part 2

Before I begin tonight’s piece, one small thing that I want to point out from my last piece was that though my models were a few years ahead of the life insurance industry, the two most important things that I did were:

  • Not optimize for best return, even if risk adjusted.  I gave extra weight to avoiding the downside.
  • Added in the details.  My models were entirely home-grown, and took advantage of my programming abilities to come to a sharper result.  I am not a good theoretical mathematician, but I am good at using math to solve practical problems.

The idea is model completely, and don’t ignore scenarios that could not happen.  My interest rate model had scenarios that mimicked what we actually got, though what we got was not a high probability.

-==-=-=-=-=-=-=-=–=

One of the themes that I came away with from the Denver conference last week was look through the windshield, not the rearview mirror.  But much of the investment industry, and retail investors are destined to  look through the rearview mirror.  I, as an actuary, have been unfairly accused of driving life insurance companies through the rearview mirror, look at this and say that we can look through the windshield, but we have to be more than the common shlubs who are the majority of the market.

To do that we have to adopt an independent disposition, and not care about raising funds, but only earning returns for clients.  If we build it, they will come.  In one sense, it is like someone who has a beautiful singing voice, but who avoids pride, and admires his voice as if it were someone else’s voice. (Apologies/Credit to C. S. Lewis)

The main point here is to look through the windshield, and exercise intelligent independent judgment.  Analyze the situation, and figure out where there is an advantage as a businessman.  The tools of modern portfolio theory will be useless here, so ignore them.  They only sharpen/obscure our understanding of the past/present by calculating parameters that are not stable or predictive.

The only positive thing about Modern Portfolio Theory is that it sidelines a bunch of bright guys who would otherwise be competitors in the markets.  Sun Tzu would admire this tactic — getting a large portion of the opposition to become peaceniks because the expected value of the war is zero or negative.  It’s as if an economic Tokyo Rose is broadcasting that competition in financial markets is futile — in aggregate, everyone will get average performance, so why fight to get better performance?  It’s futile; give up; go home.

I would simply say there is always a decent amount of lazy investors in the market. Smart investors can get better returns through paying careful attention to what seems to offer the best returns on a forward-looking basis.

Part of looking through the windshield is avoiding noneconomic constraints.

1) Do I care where a company is located? Yes. I want a place where the rule of law is honored.  As many have commented at my blog, does that include the US?  Yes, for now.  Global diversification is important.  That said, it will be interesting to see what will happen to investments should we see tariffs, foreign exchange controls, and expropriation.  At least in you own home country you only have to deal with the “devil you know.”

2) Do I care whether a company has a large or small market capitalization?  Not now.  Even if my asset management firm grows, I will adjust my strategy to include attractive small caps at lower target percentages.  If my buying begins to affect the stock price, I will take smaller positions.

3) Do I care if a stock is “a growth stock” or a “value stock?”  No, I care more about industry and firm prospects relative to price.  I will pay up on occasion.  Still, mostly I try to buy them cheap, and it biases me toward “value stocks.”

4) Do I care about whether a stock is volatile or not?  Yes.  Stock price volatility is a sign of low creditworthiness, and usually I only buy higher quality stocks.

5) Do I care about price momentum?  Yes.  Typically, I buy companies that have strong current momentum, or poor momentum over 3-5 years.  Is it what I focus on?  No.

6) Do I care if I have an “undiversified” portfolio?  No.  I want to be in the right place at the right time on average.  Mimicking the index is a recipe for mediocrity.

7) Do I care if I am holding cash?  Yes.  I’d rather be in stocks, but will build up cash if I have to.  Cash moderates volatility in a concentrated portfolio, and allows for opportunistic purchases.

8 ) Do I care if I underperform?  You bet.  It burns a hole in my gut.  But it doesn’t make me change my methods.  It will make me sharpen my analyses.

A large part of the idea is to focus on risks, not risk.  Academics focus on univariate risk, with its simplistic math — beta, standard deviation, skewness, and all of the half-measures and ratios that stem from them.  I can’t model my methods in full, but I look at the risks in particular for each of my investments.  Every investment has to justify its existence in my portfolio independently.  I don’t do correlations; they are not reliable.

I also don’t go in for the four Carhart risk factors — beta, size, value/growth, and price momentum.  I don’t think of them as “betas,” but as “alphas.”  These are factors that can be taken advantage of when they are cheap or rich.  They are not risk factors, they are simply factors.

In closing, there has been a shift in the environment from inflation to deflation.  How does that affect investment choices?  My guess: buy well-financed companies with a low price to tangible book.  Stagflation?  In the ’70s the answer was low P/E with pricing power.

The closing segment of this series will focus on how to do statistically valid studies of investment performance.  I know at least one person who may be annoyed by what I say, but it is important to try to be fair in investment analysis, lest we lead others astray.

6 Comments

  • cdudko says:

    “Judgment” is a tricky thing… many are more comfortable with precise (but wrong) outputs from formulas rather than approximate conclusions reached through judgment. I think this leads to a dramatic underappreciation of “soft” factors like management quality and corporate culture in investing. They are not the only factors in investing, but they are factors.

    It is also an excellent intellectual cover to be able to point to the “theory” when investment results don’t go as planned. Harder to say “Yep, my judgment was off.”

    • Good comment. On average, my judgment is usually good, but boy, I have had some stinkers in my life. Part of preserving good judgment is willing to be honest with yourself about how many times you err. Humility is rewarded; overconfidence gets punished.

  • cdudko says:

    Our judgment allergy extends beyond investing, whether in our preference for rules over principles in accounting and regulation (topics I think you have discussed before) or in the form of security theater courtesy of the TSA. http://www.thestar.com/news/world/article/744199—israelification-high-security-little-bother.

  • microcap says:

    I don’t have empirical data [too lazy], but I have found that re point #1, where companies are based WITHIN the US is also important. Avoid Florida, NYC, Long Island, Utah, Texas, SoCal if possible. Midwest is best, upstate New York, anyplace that still has some emphasis on moral values versus financial ones. It’s not perfect, just another factor to toss into the “judgment” model.

  • Good additional comments cdudko and microcap

  • cold.as.ice says:

    I do AML modeling and analytics and a key Columbus mistake perhaps has some wisdom to share with modeling.

    How could he have been so far off as to think that he could sail from Spain to China over open water with the supplies on his ship? There were a number of ancient and medieval measurements of the Earth’s size available in Columbus’s day. He used data from the Persian astronomer Al-Fargani, who in the 9th century had determined that one degree of latitude spanned 56 2/3 miles.

    The problem was that Al-Fargani was measuring in Arabic miles, while Columbus himself used Italian miles, and Columbus didn’t realize the difference. Columbus was underestimating the size of the earth by 40%. This underestimate contributed to his belief that reaching the east coast of Asia was doable from western Europe with the ships of his time.

    Bottom line is he did not redo the measurements and computations. Very simple techniques for accurately measuring a degree of earth circumference have been know since before the time of Christ by the Egyptians, Phonations, Arabs and others. He could have redone the measurement for 1 degree of circumference, multiplied by 360 and determine the earth’s size in his units of measure.

2 Trackbacks