Trend-Following Filters – Part 10
1. Introduction Two previous articles, “Trend-Following Filters – Part 7” [1] and “Trend-Following Filters – Part 9” [2], examined, from a digital signal processing (DSP) [...]
Fund Selection When Borrowing Is Restricted
Once borrowing is realistically restricted, the Sharpe ratio can stop lining up with what investors actually care about: utility. This paper argues that in this constrained world, the geometric mean is a better compass.
Are U.S. Stocks Running Out of Steam? A Deep Dive into Valuations and Market Concentration
We're going to examine the market’s current concentration and valuation to better understand return expectations going forward. But reader beware; this isn’t some bold macro prediction to scare you away from sensible investing. It’s a reminder that markets move in cycles, valuations eventually matter, and history has a way of humbling even the most confident forecasts.
Trend-Following Filters – Part 9
This article examines and compares, from a digital signal processing (DSP) time domain perspective, several filters that are modeled on the assumption that the input follows a second order process, i.e., the input contains a linear trend. These filters are, by design, better able to track linear trends than some other more commonly-used filters, such as moving average, exponential smoothing, etc., which exhibit lag, or a time delay, in response to trends. Filters modeled on a second order process are commonly referred to in the technical analysis literature as “zero lag” filters.
How Tiny Price Differences Help Track Small Investors’ Trades
This article explains how researchers studied small investors' trading habits by looking at tiny price differences, called subpennies, in stock trades. They found that the current method to identify these trades isn't very accurate. By using a new approach, they improved the accuracy, helping to better understand how small investors buy and sell stocks.
Factors are global, respectable and repeatable
The propagation of factors actually reflect valid characteristics of the markets and market fluctuations.
Trend-Following Filters – Part 8
This article describes digital filters derived from time series regression models that can be used as technical analysis tools. The filters are analyzed from a digital signal processing (DSP) frequency domain perspective to illustrate their properties. Example charts of the filters applied to the S&P 500 index are also included.
Using Bayesian Solutions to Resolve the Factor Zoo
We propose a novel framework for analyzing linear asset pricing models: simple, robust, and applicable to high-dimensional problems.
U.S. Companies Have Outperformed Japanese Companies, or Have They?
While both the S&P 500 and the Nikkei indices have recently hit all-time highs, the valuation and balance sheet data we have reviewed indicate that the downside risks in Japanese stocks appear to be far less than the risks in U.S. stocks. Evidence such as this helps explain why legendary investor Warren Buffett has been buying Japanese stocks.
Creating Better Factor Portfolio via AI
Trading costs, discontinuous trading, missed trades, and other frictions, along with asset management fees can cause a shortfall between live and paper portfolios. The focus of this paper is to test an effective rebalancing method that prioritizes trades with the strongest signals to capture more of the factor premia while reducing turnover and trading costs.
Complexity is a virtue in return prediction
Simple models severely understate return predictability compared to “complex” models in which the number of parameters exceeds the number of observations.
The Performance of the Hedge Fund Industry
New research reveals that the performance of the hedge fund industry has not been as bad as the results from studies that relied on hedge fund data providers.
Tracking Error is a Feature, Not a Bug
The benefits of diversification are well known. In fact, it’s been called the only free lunch in investing. Investors who seek to benefit from diversification of the sources of risk and return of their portfolios must accept that adding unique sources of risk means that their portfolio will inevitably experience what is called tracking error—a financial term used as a measure of the performance of a portfolio relative to the performance of a benchmark, such as the S&P 500.
Are stock returns predictable at different points in time?
For many benchmark predictor variables, short-horizon return predictability in the U.S. stock market is local in time as short periods with significant predictability (“pockets”) are interspersed with long periods with no return predictability.
Trend-Following Filters – Part 7
This article examines four digital filters commonly used for trend-following: moving average linear weighted moving average exponential smoothing time series momentum
Geographic Investing and Company Operations
The article explores the limitations of traditional country-level stock market indexes that are constructed based on the domicile of issuing firms.
Regression is a tool that can turn you into a fool
Running regressions on past returns is a great tool for academic researchers who understand this approach's nuance, assumptions, pitfalls, and limitations. However, when factor regressions become part of a sales effort and/or are put in the hands of investors/advisors/DIYers, "the tool can quickly turn you into a fool."
Submergence: A Tool to Assess Drawdowns and Recoveries
According to research by the authors, stocks and bonds have been submerged for about 75% of the time since 1980; and treasuries have been submerged 80% of the time. Submergences are therefore both commonplace and significant, which means that handling them is very important for investors and their investing strategies.
Does diversification always benefit investors? No.
This article examines the extent to which these assumptions hold and the extent to which investors should want them to hold. The authors deliver a clever quote from Mark Twain (or maybe it was Robert Frost) that nails the issue in simple terms: “Diversification behaves like the banker who lends you his umbrella when the sun is shining but wants it back the minute it begins to rain”. Nicely expressed!
You Thought P-Hacking was Bad? Let’s talk about “Non-Standard Errors”
Most readers are familiar with p-hacking and the so-called replication crisis in financial research (see here, here, and here for differing views). Some claim that these research challenges are driven by a desire to find 'positive' results in the data because these results get published, whereas negative results do not get published (the evidence backs these claims).
But this research project identifies and quantifies another potential issue with research -- the researchers themselves! This "noise" created by differences in empirical techniques, programming language, data pre-processing, and so forth are deemed "non-standard-errors," which may contribute even more uncertainty in our quest to determine intellectual truth. Yikes!