1/ Robust Beauty of Improper Linear Models in Decision-Making (Dawes)

"Even improper linear models are superior to clinical intuition when predicting a numerical criterion from numerical predictors. In fact, unit (i.e., equal) weighting is quite robust."

uwe-mortensen.de/DawesRobustBea…
2/ "In proper linear models, predictor variables are weighted such that the resulting linear composite optimally predicts some criterion of interest; examples of proper linear models are standard regression analysis, discriminant function analysis, and ridge regression analysis.
3/ "Research summarized in Paul Meehl's book on clinical vs. statistical prediction—and a plethora of research—indicate that when a numerical criterion (e.g., graduate GPA) is to be predicted from numerical predictor variables, proper linear models outperform clinical intuition.
4/ "Improper linear models are those in which the weights of the predictor variables are obtained by some nonoptimal method; for example, they may be obtained on the basis of intuition, derived from simulating a clinical judge's predictions, or set to be equal."
5/ "People are important. The model may integrate the information in an optimal manner, but it is always the individual (judge, clinician) who chooses variables. Moreover, the human judge knows the directional relationship between the predictors and the criterion of interest."
6/ "In multiple regression, b weights are notoriously unstable; the ratio of observations to predictors should be as high as 15 or 20 to 1 before b weights, which are the optimal weights, do better on cross validation than do simple unit weights."
7/ "Bootstrapping models use weights derived from judges; because they are not derived from the relationship between the predictor and criterion variables themselves, the resulting models are improper. Yet they consistently do better than the judges from which they were derived."
8/ "Random linear models perform about as well as the paramorphic models of the judges; these averages are presented in the third column of the table. Equal-weighting models [after standardizing the variables], presented in the fourth column, do even better."
9/ "Linear models are robust over deviations from optimal weighting. In other words, the bootstrapping finding, at least in these studies, has simply been a reaffirmation of the earlier finding that proper linear models are superior to human judgments."
10/ "People who argue that different criteria/judges/variables/time frames would produce different results have had 25 years in which to produce examples and have failed to do so.

"The illusion of good judgment may be reinforced when predictions have self-fulfilling effects."
11/ "Statistical prediction, because it includes the specification of exactly how poorly we can predict, bluntly strikes us with the fact that life is not all that predictable. Unsystematic clinical prediction, in contrast, allows us to hold to a comforting illusion."
12/ "The GRE is a carefully devised test. Do we really believe we can do a better job by a 10-minute folder evaluation or half-hour interview? Such cognitive conceit is unethical, especially given the complete lack of evidence that we do better than the linear equation does."
13/ Related reading:

Anomalies and News
"Analysts are normally wrong stocks' expected returns. Over pessimism/over-optimism could contribute to anomaly returns."


Superforecasting


Expert Political Judgment

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Darren 🥚 🐣 🕊️

Darren 🥚 🐣 🕊️ Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @ReformedTrader

27 Sep
A few real estate investors made a lot of money in 2009 and 2010, but they didn't do it because prices rebounded quickly from a low. (In the U.S., prices peaked in 2005 and went into a six-year downtrend.)

They did it by buying houses at large discounts of 30% to 50% + repairs.
There is momentum in real estate like in everything else: trends from last year tend to continue.

But even where we do see mean-reversion (one-month time frames in individual stocks), it takes a portfolio of hundreds of positions to capitalize on this statistically weak effect.
I don't know of any trading strategy that can generate reliable returns by trading a single, individual asset (one stock, one house, etc.) using its past price data.

Real estate investors get around the lack of diversification by getting huge discounts.
Read 18 tweets
26 Sep
1/ US Inflation and Global Asset Returns (Dai, Medhat)

"While average real returns were lower in years with higher inflation for most assets, many of the differences are not statistically reliable, especially among non-bond assets & in more recent times."
papers.ssrn.com/sol3/papers.cf… Image
2/ "Our conclusion that most asset classes have limited inflation-hedging abilities is consistent with the literature. Bodie (1976), Fama & Schwert (1977), & Fama (1981), among others, find that nominal stock returns are negatively related to expected & unexpected U.S. inflation. Image
3/ "Gultekin (1983) & Beckers (1991), among others, find similar evidence outside the US. Fama & Schwert (1977) also find (i) that nominal returns to government bonds and bills are only positively related to expected inflation and
Read 17 tweets
24 Sep
1/ Betting Against Quant: Examining the Factor Exposures of Thematic Indices (Blitz)

"Investors in thematic indices trade against quants, who prefer stocks that are currently cheap & profitable. Negative factor exposures imply low expected returns."

papers.ssrn.com/sol3/papers.cf… Image
2/ "Our sample includes all S&P and MSCI thematic indices with at least 3 years of data as of end April 2021.

"Our conclusions should not be generalized to thematic investing in general, since our analysis is exclusively based on data from two index providers." ImageImage
3/ "The history has backfilled data, as the first S&P thematic indices were launched in 2016; MSCI indices were launched as recently as 2020. Backfilled returns are probably biased upwards (e.g. survivorship bias), but this is less of a concern for estimating factor exposures."
Read 12 tweets
23 Sep
Return Stacking: Strategies for Overcoming a Low Return Environment (@RodGordilloP, @choffstein, @GestaltU)

"We describe a way to stack returns by choosing alternative fund managers already engaging in non-correlated, capital-efficient strategies."

info.rcmalternatives.com/return-stacking ImageImageImageImage
"How do you endure the ‘line item risk’ of alternatives? How do you participate in the upside of an increasingly overvalued stock market? What value do bonds bring at the zero bound? The answer boils down to some new capital-efficient ETFs & mutual funds."
podcasts.apple.com/us/podcast/res…
Read 5 tweets
9 Sep
1/ Explaining the Recent Failure of Value Investing (Lev, Srivastava)

"We identify two reasons for the failure of value investing: (1) accounting deficiencies and (2) fundamental economic developments which slowed down mean reversion of value & glamour."

papers.ssrn.com/sol3/papers.cf… Image
2/ "The value strategy had already lost much of its potency in the late 1980s and yielded negative returns in the 1990s, barring a brief resurgence in 2000-2006.

"The expensing of intangibles started to have a major effect on book values and earnings in the late 1980s." ImageImage
3/ "The effect of our intangibles book-value adjustments are more pronounced for glamour than for value stocks. Among glamour stocks, our adjustments had a larger effect on small than large companies, since small, high-growth glamour firms tend to invest heavily in intangibles." ImageImageImageImage
Read 10 tweets
4 Sep
1/ Is U.S. housing in a bubble?

TLDR: Maybe. Valuations are consistent with this, especially considering real yields, but borrowers are also in better shape than in 2006.

Maybe build a one-sheet Excel model for expected housing returns and decide for yourself.

Thread
2/ Most of the realtors I've talked to tell me that the housing market is even hotter than in 2005, at least when metrics like months of inventory, days on market, and rapid price increases are concerned.

3/ Where I live, houses are on the market for <20 days on average, and buyers pay an average of 10% above asking and waive their inspection contingencies. Some are coming in with all cash.

Inventory (valuation) matters a lot in the short (long) term.

Read 16 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(