Multifactor investing

Ask your investment, budget, and other money related questions here
Post Reply
wolf
Posts: 1015
Joined: Fri Jan 06, 2017 5:09 pm
Location: Germany

Multifactor investing

Post by wolf »

Does anyone have experience with multifactor investing?

My idea is to invest 20% of my portfolio with a multifactor investing strategy.
Specifically, I‘d like to invest into large US equity. I don‘t want to by an index ETF, I‘d want to invest actively into stocks and pick them by myself.

In order to invest by factor you need the necessary data to analyze. But that's difficult.

So I came up with an idea, a shortcut one could also say. What if I‘d use the available data from ishares. You could find all components of their indexes on their website.

So I downloaded the data of the following indexes here
  • MSCI USA Quality
  • MSCI USA Value
  • MSCI USA Momentum
Then I combined those three factor indexes and looked which stocks are listed in at least 2 out of those 3 factor indexes.
Then you could further select, filter, etc. e.g. by sector, market size, etc.
I'd need to review it (semi-)annually, rebalance and sell and buy.

Those factor indexes are based on the MSCI USA, which holds 637 public traded stocks. (MSCI source)

I have to confess. In regards to the CCCCCC model such an approach is probably located (only) at the third stage, compiling.
Nevertheless, it would be a viable approach of multifactor investing without buying an ETF index.

What do you think?

User avatar
Dream of Freedom
Posts: 475
Joined: Wed Aug 29, 2012 5:58 pm
Location: Nebraska, US

Re: Multifactor investing

Post by Dream of Freedom »

Why not just use a stock screener like finviz? Which criteria do you want besides large cap?

wolf
Posts: 1015
Joined: Fri Jan 06, 2017 5:09 pm
Location: Germany

Re: Multifactor investing

Post by wolf »

I look into it. Thanks for the suggestion. Haven't heard about it before.
Criteria would be: P/E, P/B, LT Debt/Equity, RoE, etc.

IlliniDave
Posts: 2784
Joined: Wed Apr 02, 2014 7:46 pm

Re: Multifactor investing

Post by IlliniDave »

Multifactor investing as I understand it is just another method of grouping the the same sorts of assets together. So right off the bat I don't know of any major benefits to doing it with a small slice. That said I do exactly that for amusement more than anything, although I stick with funds.

Your idea of looking at the portfolios (or underlying indices) of factor and multi-factor funds out there is a reasonably good starting point. It would also be interesting to compare what you might find in a screener to what funds actually hold.

ToFI
Posts: 132
Joined: Thu Jun 16, 2011 1:22 am

Re: Multifactor investing

Post by ToFI »

What are you trying to achieve with Multi factor investing ? If you want higher return, invest in fast growing companies. e.g. RingCentral(30% growth), Adobe (23% growth). If you want less volatility, increase cash/bond allocation.

I also don't get why people want to blend their mortgage with both fixed rate and variable rate. I never mix variable rate with fixed rate. Can't decide where to go ? One foot forward and one foot backward. Indecisive people won't get far in life.

I am already FIRE and my objective in investing is simply maximizing return while minimizing risk. My strategy is invest in a large numbers of high quality growth stocks with at least 15%+ growth rate.

wolf
Posts: 1015
Joined: Fri Jan 06, 2017 5:09 pm
Location: Germany

Re: Multifactor investing

Post by wolf »

Yes iDave, I'll do that. That's kind of a check.

@ToFi: I'll want to achieve the following things:
I don't want to invest into an index fund.
I want to achieve a better sharpe-ratio compared to the index, in this case msci usa.
In another way I'll want the same as you mentioned: maximizing return while minimizing risk (in this case volatility).
Of course there are different possible ways to achieve that. Multifactor Investing is one in my opinion.
Btw: my portfolio holds also other positions with other factors, than I mentioned in my first post.
So overall, I think of a equity portfolio which holds positions with different underlying factors, like momentum, quality, value, size, yield, political risk (emerging market). I want to build a equity portfolio which has as much diversification (in regards to lower volatility overall) as possible without minimizing the return.

ToFI
Posts: 132
Joined: Thu Jun 16, 2011 1:22 am

Re: Multifactor investing

Post by ToFI »

I am not fan of value investing. Stocks are cheap for bad reasons in normal market. Yes, when the stock drops, it can overshoot a little bit, that's where the price / value mis-match creates opportunity but it only exist in short period of time. It's a lot of work to get inconsistent return. The true value stocks only appear in bear market such as in 2000 and 2008.

I am also not fan of momentum investing which only looks at the price movement alone.

I think growth is best way of investing. It's boiled down to what Charlie Munger said: It's good idea to buy quality business at a fair price.
What's high quality growth stock? It has consistent above average revenue growth( 15% to 60% per year). It has a stock chart that goes up smoothly and steadily over long period of time. It has strong business model so it can continue the high rate of growth.It also helps if we focus on small and mid cap companies because large caps have limited long term growth potential.

I can pick the best stocks from a growth ETF. Set equal weight.Reason is no need to guess which stock is the best. No re-balancing. The reason is to avoid selling winners and buying losers. The result out performs the growth ETF itself in up and down market by a big margin. It suprised me that a large group of growth stocks drop less than the index. I think a highly diversified equal weight growth portfolio has multiple competitive advantages from different sectors while index is market cap weighted concentrated in a small number of stocks. In a typical index, the top 10 stocks make up of 50% of the value. . For instance, the SP500 has high exposure to 5 big tech companies. If they slow down, the SP500 will suffer. My portfolio, each stock is less than 1% of the portfolio. Yet, the past 5 years annual return was 25% per year from back testing. I constructed last month.This year return to date is 14%.
Last edited by ToFI on Sat Feb 15, 2020 1:50 pm, edited 1 time in total.

HalfCent
Posts: 40
Joined: Mon Dec 23, 2019 8:56 pm

Re: Multifactor investing

Post by HalfCent »

The result out performs the growth ETF itself in up and down market by a big margin.
Can you explain? Are you saying because you won't sell unnecessarily but the ETF managers would?

HalfCent
Posts: 40
Joined: Mon Dec 23, 2019 8:56 pm

Re: Multifactor investing

Post by HalfCent »

Also, how many stocks are you talking, @Wolf? Seems like the sheer numbers are where the index has an advantage.

ToFI
Posts: 132
Joined: Thu Jun 16, 2011 1:22 am

Re: Multifactor investing

Post by ToFI »

The categorization of Value, Quality, Momentum just create more noise.
what makes the stock goes up is simply how fast and consistent the revenue/profit are growing
Don't think all stocks listed in ETF are good stocks. It's a mixed bag. There are companies with flat revenue and stock price going nowhere for years in ETF. You should check both revenue/profit growth rate and stock performance before you buy any of them.

wolf
Posts: 1015
Joined: Fri Jan 06, 2017 5:09 pm
Location: Germany

Re: Multifactor investing

Post by wolf »

I talk about 20 to 30 stocks

Forskaren
Posts: 166
Joined: Sat Nov 07, 2015 4:04 pm

Re: Multifactor investing

Post by Forskaren »

I don’t see how you can get a good diversification compared to the global stock market with only 20-30 stocks. Also, if you mix too many strategies you may end up with something behaving like index, but less diversified.

Seppia
Posts: 1306
Joined: Tue Aug 30, 2016 9:34 am
Location: Italy

Re: Multifactor investing

Post by Seppia »

Dream of Freedom wrote:
Fri Feb 14, 2020 1:29 pm
Why not just use a stock screener like finviz?
Thanks!
I had been looking for something like this for a few years, not sure why I've never asked.
Any other resources you'd recommend?
Thanks

wolf
Posts: 1015
Joined: Fri Jan 06, 2017 5:09 pm
Location: Germany

Re: Multifactor investing

Post by wolf »

That's the tradeoff. Either I want good diversification by investing into the US stock market, or I'll try picking stocks and try to outperform the US market. Of course the risk is to underperform the US market as well. In this case (the mentioned 20-30) stocks I refereed to the US stock market.
I wouldn't call different factors a different strategy. The overall strategy would be factor investing, meaning there are positions which represents different factors.

trailblazer
Posts: 112
Joined: Sun Jan 03, 2016 8:18 pm

Re: Multifactor investing

Post by trailblazer »

Alpha Architects has some great research/articles on value and momentum. Their work seems to be academically rigorous and avoids hype. I've read several of their more detailed books as well about quantitative value and momentum.

They offer several value and momentum ETFs for both US and International where you can see their top handful of picks for each strategy (QVAL/QMOM/IVAL/IMOM). They take a highly concentrated approach (only picking the 40 "best" stocks for each strategy) compared to funds that essentially just take the "value half" of the market.

I recall they have a free screener that is quite robust in terms of identifying value and momentum factors, but it's been a while since I accessed it.

BlueNote
Posts: 493
Joined: Sat Jun 08, 2013 6:26 pm
Location: Toronto, Canada

Re: Multifactor investing

Post by BlueNote »

One of the basic principles of investing using factors is basically: exposure to a factor is exposure to a factor. The factoring math is just looking at some empirical data and doing some fancy regressions, or some such thing (I'm not that smart), to tell you how a security behaved relative to some "factor". An index ETF/Mutual fund product should also provide more diversification and should be less risky all else the same. Also a good ETF, preferably an index product IMHO, would be ensuring that you're constantly exposed to the factor by making relevant portfolio adjustments saving you time and effort of doing it yourself.

There's a great book called what works on wall street that provides tons of cool stock screens for small portfolio building. You'll get factor exposure too.

JCD
Posts: 34
Joined: Sat Jul 20, 2019 9:12 am

Re: Multifactor investing

Post by JCD »

wolf wrote:
Fri Feb 14, 2020 10:18 am
  • MSCI USA Quality
  • MSCI USA Value
  • MSCI USA Momentum
Then I combined those three factor indexes and looked which stocks are listed in at least 2 out of those 3 factor indexes.
There is a critical error in taking this approach assuming you are interested in evaluating stocks by their specific factors, particularly if you are interested in smaller stocks. iShare/Blackrock is one of the biggest index fund managers on earth. They must buy shares of everything in their index, based upon the legal weighting system they use (often by market cap). So if a single owner owns 70% of the stock and isn't selling (like a founder), that means that the indexes have to fight over the last 30%. Often these are smaller companies, with low market caps. What happens if that company doesn't have enough shares to go around. What if literally they go to the well...er... market and it is dry? Now they are forcing the price up just due to a lack of available shares. So what they do is they "weight" or index the stocks based upon market cap minus insider ownership. Then they throw out anything without enough shares based upon a model of index size expectations.

If you assume insider ownership is a factor, then you are actually not participating in that factor. Even if you aren't convinced that is a factor, there is a good chance that stocks with large inside ownership have low index ownership and thus are more likely to be impacted by "irrational activist markets forces" over the buy and hold indexer. That is to say, from my investigations, they tend to be cheaper. You may decide they are higher risk too, since the index won't save your share price, but that is a personal judgement.

I believe there are other limits to using a indexer list, like "value" maybe just the stocks that are large cap (thus indexable) but below the 50%'tile regarding pe ratio or pb ratio. That may not actually be value, particularly once you throw out any small/microcap companies that the indexers can't easily index. Again, depending on your goal, this maybe desirable, but you have to know what you are getting into.

Lastly, keep in mind that A && !A (A and not A) can both be true in index land. For example:
https://www.ishares.com/us/products/239 ... growth-etf
and
https://www.ishares.com/us/products/239 ... -value-etf

both have JPM in the list of companies they include in their respective portfolios. Now maybe JPM is both growth and value. But if you think those are opposites, then it seems pretty clear something is wrong. This is due to the need to have huge companies in the index in order to get enough shares to make the index fund viable at scale.

I would be very careful using index fund company lists for anything. Use one of the screeners, be it the one vanguard offers or finviz or one of the many others that exist (I find https://fintel.io/screener to be interesting in that they have an entire query language and like https://www.dripinvesting.org/tools/tools.asp because it is in excel). If you are shooting for value, I would suggest going beyond just factors and examining the books and reading the earning call transcripts. Its a lot of work, but the risk of accepting the judgement of historical(*) factors, other people's opinions and industrial sized indexing systems is pretty high.

(*) Example: P/E using past 12 months earnings, not future earnings. Yes PEG might be good, but it is just an estimate. If you don't know how that estimate came to be, you are trusting other people's opinions.

Post Reply