All that glitters is not gold: Comparing backtest and out-ofsample performance on a large cohort of trading algorithms

Jim,

I am glad you find the ideas in my ETF sector rotation model useful. I have actually test the rotation model with ETFs from different providers and sticking with Blackrock (i-shares) ETF universe seems to yield the best result. (definitely exclude the Vanguard ETFs since their performance are alway inferior). For sector rotation ETF strategy,it may not be too relevant to test it in different markets since the economic conditions (including the growth in different sectors & industries) are not the same for US, UK, Europe and the rest of Asia.

Regards
James

James,

You started this post with extremely relevant ideas that were criticized. Marco did find the article useful and commented favorably, however. That is how he built a successful business—by being open to new ideas.

And not nit-picking the details of every good idea people try to share with him. At least I have never seen him do that.

I don’t think everyone else did as well as Marco, including me. I just like all of your ideas it seems. But I am sincere in that. Time will tell if–as good as he is–Marco might be missing some useful ideas.

I am going to leave the last ideas in this thread to you (for now anyway). Maybe I will think of something important that you have not already mentioned. I am sure others will have useful comments.

But again extremely relevant ideas.

Excellent. No surprise considering you training and experience.

Thank you.

-Jim

Georg,

After spending the past week building stock factors on different ETFs, mutual funds and hedge funds, it is now clear to me that it is way better to ride on the performance of fundamental star hedge fund managers than ETFs or mutual funds with good past performance. I have now chosen four hedge funds (Tiger Global, Bluecrest Capital, Lone Pine and Viking Global) to replicate and will allocate capital in these strategies. It seems that Goldman Sachs is doing more or less the same and provides these portfolios for their institutional clients (pls see below the two article from CNBC).

Regards
James

Goldman has a portfolio which tracks hedge funds that is trouncing the market — here’s what’s in it

Published Fri, Feb 22 201910:59 AM ESTUpdated Fri, Feb 22 20191:45 PM EST

Key Points

: The so-called hedge fund “very important position” basket contains the 50 stocks that appear most often on the top 10 holdings of fundamentally-driven hedge funds.

: The portfolio is up 14 percent year to date, more than doubling the average equity hedge fund’s 6 percent gain and outperforming the S&P 500′s 11 percent return.

: Goldman revealed that the most-owned holdings in the basket are very tech heavy, with Amazon, Microsoft, Facebook, Alphabet and Alibaba being the top five.

: The basket also has a track record of beating the market as it has outperformed the S&P 500 in 62 percent of quarters since 2001, Goldman said.

One of Goldman Sachs’ secret portfolios that tracks hedge funds’ most popular long positions is crushing the market.

The so-called hedge fund “very important position” basket contains the 50 stocks that appear most often on the top 10 holdings of fundamentally-driven hedge funds, according to the firm. Goldman analyzed 880 hedge funds with $2.1 trillion of gross equity positions to compile the latest portfolio, which is based on funds’ recently released fourth-quarter regulatory filings.

The portfolio is up 14 percent year to date, more than doubling the average equity hedge fund’s 6 percent gain and outperforming the ’s 11 percent return. Goldman revealed that the most-owned holdings in the basket are very tech heavy, with Amazon, Microsoft, Facebook, Alphabet and Alibaba being the top five. These five stocks also made last quarter’s top five.

“Recent hedge fund returns have benefited from the outperformance of the most popular long positions as well as the decision to increase net length ahead of the equity market bottom in December 2018,” Ben Snider, an equity strategist at the bank, said in a note Friday.

The smart money managed to chase the huge tech comeback with these top-five names rebounding from their December lows and rising as much as 25 percent in the new year. Social media giant Facebook is up 22 percent in 2019 after losing more than 20 percent amidst fourth quarter’s market turmoil. E-commerce powerhouse Amazon has posted a nearly 8 percent gain in the new year after losing more than 25 percent in the fourth quarter. Alibaba has returned more than 25 percent year to date.

The basket also has a track record of beating the market as it has outperformed the S&P 500 in 62 percent of quarters since 2001, Goldman said.

Other stocks in the hedge fund VIP basket include Visa, Netflix, Bank of America, Paypal and Citigroup.
Rising Stars

By the end of the fourth quarter, hedge funds had ramped up their bets on CVS Health as 108 out of 880 hedge funds held the drugstore chain, up from 64 fund last quarter, the largest increase in hedge fund popularity, according to Goldman Sachs.

Software company Red Hat also gained popularity with 83 hedge funds owning the stock, up from 42 last quarter. Notably, Warren Buffett’s Berkshire Hathaway last week revealed a new stake of 4.175 million shares in Red Hat, according to the company’s 13-F filing.

Although gaining popularity, CVS and Red Hat still didn’t make Goldman’s hedge fund VIP portfolio because they’re not yet in the top holdings of enough funds.

A few of the secret portfolios Goldman gives clients are doubling the market’s return this year

Key Points

: Three of Goldman’s secret portfolios are posting a 11 percent gain so far this year, more than doubling the S&P 500′s year-to-date return.

: A majority of Goldman’s 39 proprietary baskets are beating the market this year.

: These portfolios are only made available for its clients.

: The three top baskets this year are “high revenue growth,” “high Sharpe ratio,” and “dual beta.”

Goldman Sachs apparently knows the key to beating the market, but the bank is only telling its elite club.

The bank has created 39 portfolios exclusively for its clients, and 34 of them are outperforming the market this year while three are posting a 11 percent gain, more than doubling the ‘s year-to-date return. The three winning baskets are “high revenue growth,” “high Sharpe ratio,” and “dual beta,” according to a broad note to clients on Monday explaining the portfolios’ constructions and growth outlooks.

Goldman has been handpicking stocks based on various proprietary themes and sectors for a decade, but these portfolios are only made available for its clients to trade on the Bloomberg terminal. The bank treats the members of the baskets carefully and doesn’t always disclose all the member stocks, even in its daily and weekly notes.

Along with good track records this month, the baskets also have solid long term track records. However, the bank doesn’t suggest buying and holding the various baskets all at the same time. They recommend certain baskets at certain times based on the type of market environment.

For example, Goldman warned clients at the end of last week about companies with big revenues from China, and specifically Nvidia, days before the chipmaker released surprise disclosure on China slowdown and blew up its stock. In the note, Goldman told clients to avoid its international sales basket and recommended its domestic sales basket instead for investors wanting to hedge against international risks.

High revenue growth

One of the best-performing baskets this year consists of 50 companies in the S&P 500 that have the highest expected sales growth based on the Street’s consensus. The equal-weighted portfolio has returned 10.7 percent so far in January and 143.4 percent since its inception in May 2011, versus the S&P 500′s 5.4 percent and 137.3 percent respectively.

“This basket focuses on companies positioned to use top-line revenue generation instead of margins to drive bottom-line earnings,” Goldman’s chief U.S. equity strategist David Kostin said in the note.

It’s actually no surprise that this portfolio has performed well this year as fewer and fewer companies are able to continue drive organic sales growth on the heels of a global economic slowdown. Many companies have voiced concerns this earnings season about the impact from the slowing demand and weaker consumer confidence.

Netflix, Align Technology, Amazon, Autodesk are among the dozens of stocks in this winning basket.

High Sharpe ratio

Another winning factor this year is the Sharpe ratio, a measure of a stock’s performance relative to its volatility. Goldman uses consensus price targets and options six-month implied volatility to measure Sharpe ratios.

“The inclusion of a risk metric has led to consistent outperformance on an absolute basis. The median stock in our basket offers almost double the expected return than the median S&P 500 stock with similar risk,” Kostin said.

The portfolio has returned 11.5 percent year-to date and a whopping 238.8 percent since its inception in December 2009. Stocks selected in this basket include PVH, Conagra Brands, Western Digital and Assurant.

— With reporting by Michael Bloom

I have an Asset Manager model that does the same.
You can follow the live performance here:
https://www.portfolio123.com/app/r2g/summary?id=1589757

Georg,

I just checked out your link, the model you put together invests in asset managers.

The hedge fund model that Goldman Sachs put together invests in the 50 most commonly held positions by a big list of hedge funds. It is not the same.

Regards
James

James,
No it is not the same, but the performance is similar.

Thank you for your post and the links. There is valuable information there for the piggy-back strategy.

Georg,

This is the one of the reasons why I choose 3 tiger funds (Tiger Global, Lone Pine and Viking Global) to replicate.

Regards
James

Hedge Fund ‘Cubs’ Share Information — And Alpha

Research shows that hedge fund “families” — like Tiger Management and its offshoots — make overlapping trades that result in excess returns.

Amy Whyte

February 14, 2020

Research suggests that portfolio managers who have left hedge funds to start their own firms continue to benefit from those old ties.

Hedge fund “families” — think Julian Robertson’s Tiger Management and its assorted Tiger Cubs — likely share information or ideas among themselves, according to the study by two finance professors at the University of Hawaii at Manoa. Nimesh Patel and Harold Spilker documented overlapping trades by hedge funds with shared origins, and found that a portfolio of these coordinated positions resulted in excess returns.

“This is surprising given the lack of formal connections between independent hedge funds,” Patel and Spilker wrote. “Why would fund managers, each running legally independent funds and competing for flows, share investment information or ideas?”

For one, most hedge fund managers lack the capital to push market prices up or down by themselves, the authors suggested. Managers also may share information to elicit feedback or find out what others know, as previousstudies have concluded.

Whatever the motivation, Patel and Spilker found that information sharing among hedge fund families “manifests as family block trades that predict stock level alphas.” Specifically, a long-short portfolio made up of coordinated family trades was found to have an annualized alpha of 7.32 percent.

To identify hedge fund families, Patel and Spilker used a dataset from Novus Partners that connects firms based on managers’ past employment history. David Stemerman’s Conatus Capital, for example, is a child fund of Steve Mandell’s Lone Pine Capital — in turn an offspring of Tiger Management.

They then analyzed “unanimous” trades made by hedge fund family members — for example, when two or more family members bought a stock that none held previously, or when each family member holding a particular stock exits that position entirely.

“The risk-adjusted returns are realized over the first two months after portfolio formation and do not fully reverse,” the authors wrote. “This suggests hedge fund family trades may contain information on fundamentals rather than short-term price pressure.”

Patel and Spilker added that this information is “likely not tradeable for other market participants,” because the returns are realized before the publication of 13F regulatory filings that report hedge fund stock positions.

“The returns are larger for high information asymmetry stocks and suggest that hedge fund families coordinate on information, despite having no shared legal structure like mutual fund families,” the authors concluded.

James, are you grouping the holdings of the three funds into one strategy, or are you running three different strategies?

Georg,

I am running three different stratgies but will not double/triple on the overlap names.

Regards
James

Jim,

I just checked out the Designer Model pages and realized that only 4 Designer Models have execess return of more than 10% for the past 1 year + 2 years.

Out of these 4 Designer Model, only 2 has positve excess return during the last 3 months despite the decent rise of the US market in the last quarter.

This shows that the nice equity curves and favourable backtesting results (that so many P123 members are focusing on) is not a good predictor of future returns since these simulated performance can not be replicated with live data. I think there is an urgent need for Portolio123 to consider providing some kind of out-of-sample performance measurements.

Regards
James

James,

Is it 12% that are beating the benchmark over the last 2 years?

For me personally this is useful information.

Do I think I am so good that I can beat 88% percent of the very smart designers and beat my benchmark?

Honestly, I have to be realistic. Probably not. This gives me a good idea as to how much confidence I should have in my own models. This is useful information.

The designers can make any decisions about their own models or designer models in general without me. I do not have any designer models.

-Jim

Jim,

I think most designer models are probably value oriented strategies. I can see in some of my own live ports that quality and low volatility are outperforming without regard to valuations. Those strategies are strong in utilities and tech. Neither of which are value sectors. So factor investing is working if you are in the right factors right now.

Jeff

Absolutely true.

And I do not know when or if it will happen but those value models will pop after any significant downturn, I think.

-Jim

Jeff, which is the right factor now. I don’t think anybody knows.
But avoid all Vanguard factor ETFs please.

Why Vanguard Should Retire The U.S. Momentum Factor ETF
https://imarketsignals.com/2020/vanguard-retire-u-s-momentum-factor-etf-vfmo/

Jeff,

Georg definitely has made a point here. There will always be some factors that works in any given timeframe.

If you want to know the recent performance of factor investing, just look at the returns at AQR during the last 2 years . Its AUM has fallen by more than 35% during this period. Even Cliff Asness himself says it is a crappy time for factor investing accoriding to this article from Bloomberg.

https://www.bloomberg.com/news/articles/2019-05-15/aqr-s-asness-is-right-it-s-a-crappy-time-for-factor-investing

Regards
James

I stay away from factor ETFs. I do think momentum can work if paired with cheap valuation, longer term underperformance, and low volatility. Basically you try to ride the mean reversion back out of the trough. I question how well pure momentum works. I have never been able to produce reasonable returns in backtesting with most momentum strategies.

Jeff

One can do a quick backtest on P123 to check which of the ETF providers does best, because if one holds all four factors (Momentum, Value, Quality, and min Volatility) equal weight the return should be the same as for the benchmark VTI.

iShares does best, the 4 factor funds match 100% the performance of VTI.
Fidelity is marginally lower.
Vanguard is a disaster.

Aren’t your expectations a bit high, sir? The number of people who have made an excess return of more than 10% in each of the last two years has got to be pretty tiny, especially when 2019’s benchmarks all made over 30%.

Personally, I’m totally fine with the fact that I massively underperformed the S&P 500 in 2019. The S&P 500 was nutso. I beat the market handily in 2016, 2017, and 2018, and I have every expectation of doing so again in 2020, unless it’s another year like 2019, in which case I’ll still make plenty of money. And because of the magic of compounding, I made more money in 2019 than I did in 2018 or 2016, even though I would have made twice as much investing in SPY.

As for designer models, some of them did just fine in 2019, and some of them didn’t. Just like mutual funds and hedge funds and ETFs.

Just keep things straight. How many people have outperformed the market by 10% year after year? Warren Buffett, Shelby Davis, George Soros, David Einhorn, Peter Lynch, Charlie Munger, Stanley Druckenmiller, Jim Rogers, and Joel Greenblatt all had AVERAGE outperformance of greater than 10%. But year after year, it gets harder. In some years, I would guess that almost none of them beat the market. Look at Buffett’s outperformance. Over the 55 years between 1957 and 2012, it varied between 45% and -20%; in 24 out of those 55 years, it was less than 10%. Excess returns are great, but consistent returns are better, and when markets get crazy, that means underperformance.

Yuval,

I think I must agree with you or why would I be here? I do not even have to think about it.

But then there is the question of risk. That can be controlled to some extent. If my ports ended up being more volatile than I expected then I can decrease my exposure. Still the risk is greater with ports compared to passive investing, I think.

What is the expected reward? You are right 2 years is probably enough. 5 years is probably better. 10 better still unless you think the market is changing, of course.

In medicine they hit you early, hit you hard and never stop with the question of risk/reward. For every drug, procedure and test.

I say here, at P123, it is the predicted reward versus risk, cost and time spent (opportunity cost really) going forward.

I am still running those spreadsheet calculations.

Yep. It is an ongoing calculation that can change.

-Jim

I think Georg’s analysis on the Designer Models (pls see attached) shows their relative -ve performance as compared to SPY,

Regards
James


P123_R2G_List+12-9-2019.xlsx (56.4 KB)