David - I figured out along time ago that there is a fundamental issue with how the bucket performance is calculated for RS performance graphs. If you have one really strong year, the top bucket may be 50% or more. If you have several mediocre or even negative years surrounding the one strong year, the one year can unduly influence the overall multi-year results. The best way to see this is to select a momentum ranking system and view the performance graph one year at a time. If my memory is correct, you will see complete reversals on a year-by-year basis. But the strong year (big results) will dominate the overall presentation of performance, making the ranking system look like a winner when in fact it may be a loser most of the time.
Steve, interesting. I did not know that. I will try your suggestion. You kind of alluded to this in another thread, I think, but I did not pick it up. Ill try this on a per year basis. Thank you.
I am not employed and never have been employed by big money firms, but being a participant in the market for over 10 years now, reading books, etc. etc. I would bet big that the answer is a clear “YES”. Art Cashin and Bob Pisani talked about this on CNBC today. In my opinion it is a very simple business decision to make: Spend $ on private polls and profit $$$ by being smarter then the market. In this case going long or short the British Pound is a very straightforward big money play as it can be played with high liquidity and leverage. We need to be humble on this, we are not playing in the big leagues. We are just picking up the crumbs that are left behind which in my opinion can best be found in the low liquidity part of the market where the big money does not want to play.
However, I am convinced that those who do not adept will lose their alpha. For example the same model was able to extract on average ~110% alpha out of the low liquidity market during the five year period from 1999-2004 and ~46% during the five years from 2011-2015. Coincidence? Statistically insignificant? Maybe, but I again would bet heavily that the halving of the alpha is largely because of improvements made by people like us with help of sites like P123. And what happens if we would have a user friendly program as advanced as IBM’s Watson running on P123 that would be able to create a very advanced and daily auto-adjusted model by a simple click on a mouse button? And what if that model is fully auto-traded? Maybe we would have to be happy with ~25% alpha in the same low liquidity market, under the condition that we use these advanced tools. But it would still make me happy…
Are current big hedge funds like Renaissance Technologies already using this technology? Yes, they are… but luckily they are not really interested in playing in the low liquidity market having $65 billion AUM. But I am convinced that their current technology will eventually be ours. And what will happen after that? I would guess the same ‘cyberwar’ as is currently happening in HFT (High Frequency Trading), but then spread throughout the market.
What do you mean? More quickly than humans? What about forecasting the weather? Or what about the amusing book “Trading Bases: How a Wall Street Trader Made a Fortune Betting on Baseball”. On top of that is Warren Buffet investing in IBM… for me it is a no brainer that it is related to IBM’s A.I. product called Watson. The A.I. hype is real, but difficult to grasp are its implications.
Thanks for pointing me to this pdf. I agree on the situation for which you describe jackrabbit traders as “But they work under entirely different frameworks than we do and have different tools and data sets”, if it is placed in the historical context of the early 2000’s. The landscape is changing, these jackrabbits are becoming more patient as they need new and fresh prey!
From my understanding of the book “Dark Pools: The rise of A.I. trading machines and the looming threat to Wall Street”, ‘they’ really loved to close each day with a guaranteed profit and no positions. But that almost riskless easy money deteriorated from the market since the early 2000’s. ‘Their’ current stop is closing the week or even month with a guaranteed profit and no positions. Guess what happens next?
In short, ‘they’ will hunt easy alpha first and move on to the more difficult alpha when the easy alpha has disappeared or is no longer profitable to hunt. Many of us P123 users have the exact same thinking by changing/adapting our models if they no longer are delivering the alpha we want. Also, we are naive if we think that the so called patient alpha is and will exclusively be reserved for human traders only.
True, most of these ‘users’ are media-shy because of the fear of creating more of ‘the market is rigged’ controversy then already present. The media just stands in their way to become filthy rich before it is too late and/or program the best A.I. in the world which takes an enormous amount of dedication, hard-work and IQ. Media is just the last thing ‘they’ need except maybe for spinning it saying that they are good and honest guys trying to help the market to become more efficient and liquid while taking a small profit doing this…
And what about the bloggers? They are just piggybacking on the hype as most bloggers do on any hype and most probably are a lot less educated on this subject then a simple someone like me who read a good and recent book about it.
Just to be clear: all my trading is directed by P123 “algos.” Not really a lot of AI in my ports but a pretty impressive computing and data management nonetheless. Is this lack of AI good bad or who knows? In the future no one will be required to add any AI functions that do not work. Many of us are already using trading algos through IB. Would more AI help here? I’ll leave out my opinion on this for simplicity.
I do not see the debate about algos. One could debate the importance of AI in the algos but the algos have already won control of my trading money. Some of yours too?
David / Marc / Jim - I’m not sure that pullback strategies are really contrarian, perhaps on a very short term basis but the long term trend has to be up for pullback to work.
In my opinion, contrarian strategies are the best way to go. In what is close to being a “zero sum game”, you will never profit in the long run if you are following the herd. It just doesn’t happen. You need to be first in and first out. The way to do that is by being a contrarian. But you have to make the right picks. You want to try to avoid falling knives at all possible.
As David alluded to before, you can be contrary to analysts, also you can also be contrary to short interest figures say, or contrary to institutional investors. I am more comfortable with being contrary to analysts than using a falling stock price.
I did a little bit of experimenting but didn’t have time to write it up. I used the S&P 500 universe because that is the only area that I’m truly interested in, most if not all of the stocks have analysts following them, and value factors tend notto work as well on LargeCaps.
In short I found the following were quite useful contrarian filters/rules:
FRank(“LTGrthMean”,#All,#ASC)>50
FRank(“PriceTargetMean/Close(0)”,#All,#ASC)>50
FRank(“AvgRec”)>50
FRank(“SiRatio”)>50
I tried not to optimize the rules, hence the " > 50 ".
These rules generally added a few % Alpha without overly narrowing the stock universe and with similar drawdown. Not all of the above rules work appreciably better in combination. It is probably too much of the same mechanism.
A couple of interesting things that I found:
FRank(“PriceTargetMean/Close(0)”)<10 seems to work quite well as a short selling rule. It is very peculiar that analysts do a poor job everywhere except for the lowest price targets. It could be that there is something else going on there (like N/As) but I haven’t had time to investigate.
Si%Float didn’t work as well as SiRatio as a contrarian indicator.
Institutional ownership Inst%Own didn’t work at all as a contrarian indicator. However I found that I could do better than average by avoiding the bottom 20% in terms of Inst%Own.
Most of the above rules are pretty impervious to the rebalance frequency, from 1 week right up to 1 year.
If you would like a pleasant surprise then try the following two rules in the screener with the universe set to S&P 500:
(1) FRank(“AvgRec”)>50
(2) FRank(“Inst%Own”)>20
(Hopefully I remembered the right rules
For future work, I want to focus on separating cyclical from non-cyclical stocks.
What of Brexit and Algos? I bet some of the Algos knew to ride the wave yesterday and go short at the end of the day. But of course I do not know this.
Without a doubt, as stocks bounce around today being able to respond in microseconds could be a benefit.
But as far as testing my hypothesis honestly, I do not know if there was a private exit poll yesterday and I do not know what the results were if there was one. All speculation on my part.
Since momentum and contrarian are often seen as polar opposites, a separate thread with a new heading might better attract remarks from folks who may be interested in momentum but not the contrarian topic. It’s all about target marketing!
Negative-momentum is not the same thing as contrarian view. Just to throw out an example, the common belief for some time is that long term interest rates “have to rise” as we are at or near a floor. Yet you would be making a pretty penny off of a contrarian view.
I put up TLT as a contrarian position on this thread in the hopes of some discussion. But I got none. Only condemnation on another’s graduate course. I have asked in other threads whether they believe TLT is a contrarian position. I didn’t get a Yes or NO answer but an opinion that interest rates will go up and holding TLT is not a wise move. Thus, in spite of no answer, I have to interpret TLT as being contrarian to mainstream thought. YES??
Now TLT is up 8 points since my last post. Will it continue to perform? Who knows. It is probably time for some profit taking and wait for a new entry. After all, the Fed Reserve is going to do some fist pumping that will drive TLT lower for a little while. That will present a new Buy opportunity. You see, the problem isn’t that there is a floor on interest rates, but that there is a ceiling. The US simply can’t afford to return to normal rates with a $20 Trillion debt (and counting). Raising rates would mean killing an anemic inflation rate, something the government is depending on to grow out of the debt problem. It would also kill the McJob economy, no matter how pretty picture the Fed Reserve tries to paint. Raising interest rates by even 1% would cost $200 Billion per year in debt servicing. I think even the lame Federal Government can figure out better use for that kind of money. One simple possibility would be to “buy out” the older population of workers, allowing such jobs to be held by the young millenials who are unemployed and can’t find work.
As for contrarian investing, I never thought of it as playing against the market movement, although I suppose that it is contrarian. Market movement can be due to profit taking, or general economic health concerns. On a micro or company level - Yes - it is contrarian but also dangerous unless you are truly an expert on that company and industry. When I think of contrarian investing, it is at the Macro level, and contrarian to the “expert” or establishment view of what should be the future. Playing contrarian to the Fed Reserve can be very profitable as they are either suffering from delusions or have their own agenda that is not compatible with the US public’s best interest.
Another contrarian view is health care stocks, the expert view being that they are overpriced. But this is an argument for another time. Right now I am just enjoying the renewed success of Defensive Sectors (Tax Smart) after the health care industry has taken blows from Hillary Clinton and Janet Yellen over the past year.
Presumably, this refers to Topic 8 of the Strategy Design seminar involving timing and hedging. I’m not good at integrating images with forum text, so instead of reproducing it here, I provide a link to the pdf.
And actually, it wasn’t condemnation. It was an effort to lay out the various probabilities. Feel free to address them, if you wish.
Sorry I missed it, but here goes:
Yes, TLT is a contrarian position. But that doesn’t mean it’s a good contrarian idea or a bad one. Further analysis is required for that.
Yeah Jack, you got me. Ooops, you’re not Jack. You’re Steve. Sorry about that. For a moment, I thought I was replying to the husband of my wife’s friend, the guy who sneered and wagged his finger at me because real estate continued to rise even after I argued it was a bubble and urged him not to go into that partnership for a development project in Flushing Queens. (A year later, at the end of 2008, he had lost $1 million on it.)
D’ya think!
WHOA THERE Jack . . . sorry, Steve. Actually, Jack had a better chance of being right than you do.
Yes, there is a floor and it’s a HUGE problem for the bond market. It’s called zero.
And yes, I’m aware that there is some negative-interest-rate rhetoric being bandied about. So let’s really dig into this.
Actually, I first encountered it about 10 years ago when a Federal Reserve researcher responding to an article I wrote when I was at Reuters corresponded with me and shared a white paper that resulted from a study of the topic commissioned by the Fed. As would be expected of any serious thought on the topic, it had to cover implementation. That paper postulated an “excess liquidity tax.”
Seems like a feasible answer. How else might one put through negative interest rates? Are savers/investors to purchase zero yielding Treasuries and send a fee to the Fed that allows them to hold them? How might this work in the secondary market? The Fed is still paying interest on the original face. So is there a formula that would impose a higher holding fee to more than offset the cash payments the fed makes? What are the tax rules? The current forms can’t comprehend interest as a negative; the tax software will crash the way everybody expected the rest of the world to crash at Y2K. Will we need new regs to allow the holding fee to be booked as a transaction cost? Will a new line item need to be created? What happens to corporate debt? If treasuries go low enough to give Jack/Steve the kinds of returns realized historically on TLT, as shown in sims, it’s unrealistic to think the spread will widen so far as to keep corporate rates in the black. Market forces will drive them down as people switch out of treasuries. If the corporate rate goes negative, that means banks and bondholders pay corporations a semi-annual fee for the privilege of holding zero interest-rate IRAs. This would wipe out the equity market, during all stocks to zero. Is there any corporation that won’t look to retire 100% of its equity in order to finance entirely with debt? If rates go negative enough, the holding fees can more than offset the risk of principal default, meaning even the least credit-worthy borrowers get all the capital they want.
The white paper didn’t go that far into details and the Fed tossed it onto the good-for-shits-and-giggles pile because they figured it was a political non-starter due to the clear-cut career suicide it would mean for anybody who supported it. Although the paper pre-dated the modern era, fast forwarding to today, this is one issue on which I can easily see Ted Cruz and Elizabeth Warren as enthusiastic allies. Lizze: how dare anybody try to rip-off the people by making them subsidize banks and the Fed in order to maintain their life savings? Teddy: We already have way too much in the way of taxes and complex regulations, no way we do this.
So, Jack/Steve, that’s my vision of negative interest rates. It makes for cute contrarian rhetoric. But in my opinion, this is probably the silliest view since the shmuck who paid the peak price for tulip bulbs and whose heirs are still waiting for prices to recover.
Finally, a worthy discussion .
I would agree that there is a ceiling on interest rates but would argue that we don’t really know where the ceiling is.
Current conventional economic wisdom is that as we raise rates, we depress investment. That’s pretty much what I learned in school back in the ‘70s and what I saw in the real world for a while. But what I also learned from observation is that this is not a linear function. The elasticity of the demand for capital viz. price is generally elastic but not universally so. At a certain point, the elasticity vanishes and we’ve been at that point for a long time.
I constantly dig into individual companies, both for my writings and because it’s something I do. Frankly, I have not heard one CFO cite cost of capital as a reason for not investing in more than 20 years. What I have heard, and still hear, is discouragement over revenue prospects. That’s what inhibits investment nowadays. This is why recent Fed efforts to prime the economy by pushing rates down have failed. Businesses don’t care. They want customers with money in their pockets and have long ago stopped caring about cost of capital. Those who feel they can sell continue to invest aggressively. Those that don’t have been and still are scaling back and shutting down. At some point, if/when rates rise, this will change. But given all the rhetoric and action around the obviously-wrong belief that recent efforts to slash rates would pump the economy, I suspect that economists today are flustered and scrambling, and have absolutely no idea where that ceiling really is.
At the time I was in school, the monetarists were making hay with their MV=PT model, which influenced the Regan administration in its early years but fell by the wayside as it got more entrenched, more pragmatic and less purely ideological, as so many wind up doing.
M = money supply
V = velocity of money; i.e. the pace at which money circulates
P = prices (think CPI etc.)
T = transactions; think GDP; etc.
We can measure M, P and T. The Fed can control M.
V is tricky. Early on, it was thought to be pretty-much constant. Thus you control the nominal economy by moving the money supply up and down, and watch as you go along so that P doesn’t crazily outrun T.
As it turned out, V is not constant at all. It’s been plummeting for decades and accelerating its decline recently. That’s why nothing the Fed has been doing has influenced the economy. They pump M, but V falls.
V is not controlled by any government body. It’s something that occurs naturally based on the emotions and beliefs and opinions of economic actors, mainly consumers and businesses. My belief is that decades of cost cutting, layoffs, outsourcing, etc. have walloped consumer sentiment and played a big role in driving V lower (this is part of the phenomenon that’s propelling Trump).
My opinion, I can’t prove this, it’s just a hunch, is that an increase in inflation to a non-excessive degree, say 4%-5%, would do wonders to arrest the declines in V and possibly even improve it. Objectively, consumers earn more and pay more, but I don’t see that as a problem as long as it’s more or less in balance. If consumers feel better at having more, they feel empowered over what they can control, how much and how often to buy higher priced goods and services. Emotion is important. Inflation is demoralizing only when it runs away and for that, we may have to wait a decade or more, or take a trip overseas, to Venezuela perhaps.
So again, I agree that there is a ceiling on interest rates. But in terms of how much concern we should have about it, I see us as being analogous to vertigo sufferer who is told: “Stop worrying so much about what it feels like to be on the roof deck of this high rise: You’re still holed up in the fu**ing basement!”
So if I figure out how to make a video for YouTube and post it and don’t come off as too much a dork, do promise to link to it?
Don’t have a firm opinion yet on contrarianism re: healthcare.
Going back to the subject under discussion (somehow I missed the initial posts), contrarian investing means investing contrary to prevailing sentiment. In other words, investing in stocks that few people are investing in while avoiding stocks that lots of people are investing in. Ditto with factors.
This is the cornerstone of my real-money strategy, and I’m up 24.5% YTD holding 15 to 20 stocks at a time (that’ll mean a CAGR of over 50% if 2016 continues along this route). My most heavily used factor is simply avgdailytot(90) with lower numbers being better (with some liquidity rules, of course). I also like investing in stocks with low turnover, i.e. vol/float.
The other key to being a contrarian quant investor is avoiding the most common strategies. I deliberately don’t use the TTM P/E ratio; ditto with EBITDA/EV. I deliberately pay no attention to ROE but use Greenblatt’s retun on capital instead. The accrual ratio I use rewards companies whose earnings are substantially below their operating cash flow–that’s a pretty contrarian position. I favor companies with a middling year-on-year sales growth–again contrarian. I look for stable operating margins, while most people like to see those grow. And price movements–past prices in general–play absolutely no part in my investment choices. That’s probably my most contrarian move.
Even when I apply these strategies to a smaller universe of more liquid stocks, they still work fine. I think contrarian investing still has a healthy future. There are always going to be market inefficiencies, and the whole job of a skilled investor is to exploit them.
Well Jill - Oh sorry, I just confused you for that skirt-wearing systems analyst from a year ago. I believe she warned against TLT.
For the last year the average Smart Alpha model is down 3.04% (not including non-performers that were killed by their developers). In the same period, TLT was up 27%.
Chipper quite eloquently laid out the arguments as to why TLT makes for a good hedge some time ago. If I had the time I’d dig out that post. In any case you must be quite steamed that TLT hedgers and market timing is starting to take over the Smart Alpha leader board
Marc - you are inventing terms, I have never heard of “bad contrarian” and “good contrarian”. There are mainstream positions and there are mainstream-contrarian positions. As for Jack, TLT is not an illiquid investment like real estate. You can sell TLT any time if the position starts to move against you. If you are right with a contrarian position then profits are massive because you are generally trading against the herd. If you lose, then generally losses are not as bad. This is the point of being a contrarian, not following the herd.
The point of low interest rates is to get consumers to spend. Instead consumers have started to save out of fear of the future. So like it or not, the answer is to force spending by negative interest rates. Impossible? Well we both know that 40% of the developed world has negative interest rates at present. Your girl, Janet, wouldn’t have determined if negative rates were legal if it weren’t a possibility in the USA.
Banks will survive. Tax reform is not a significant roadblock when it comes to taking money from the masses. The masses will be unhappy but who cares.
Lets put it this way, the federal reserve is between a rock and a hard place. They are shooting blanks and have been for some time. We can’t put all the blame on the fed reserve because they aren’t responsible for the huge deficits the government is racking up. But either way, higher interest rates, or lower rates, the Federal Reserve tenure is likely over. (I know you think this is another silly statement). When the Fed Reserve tries to raise rates by any significant amount, the government light bulb will turn on (well I know that sounds funny) but Mr Trump will eventually figure out that there are better ways to employ hundreds of billions of dollars than servicing debt. And if the Fed Reserve is completely ineffective, then why employ them? They are costing the country a pretty penny.
So there you go, you think my position is “silly”, but that makes for all the more profit if the trade goes well.
Well I’m long TLT and EDV and have been for awhile. The 35 year downward inflation trend is still firmly in place. Smart guys like Gary Shilling have positioned their clients at the long end of the treasury curve since the 1980s and beat the S&P 500 by 5.5 times on a total return basis. Will the next 35 years look the same? No, I think not. But for the next year or two I’m comfortable with TLT. If anyone know G.S. current position, please let us know. I no longer subscribe to his newsletter.
So the current concern is disinflation or outright deflation in Japan, Europe and the US. It’s becoming obvious that monetary policy can only do so much and without some fiscal stimulus, we’re going nowhere soon. And I don’t see a lot of fiscal stimulus happening given the worries about excessive sovereign debt.
What happens to interest rates if inflation goes negative?
What happens to the ability to service debt when future money has more value?
What happens when the velocity of money falls below one?
I don’t know but I think we’ll find out within the next 10 years.
I am not about to rehash my position (selectively) in favor of TLT. But there is one point that TLT bears keep missing: You can make money on TLT even during rising rates (depending on how fast they rise). That’s because you collect the interest while waiting.
Disclosure: I switched out of medium-duration bonds and into short-duration bonds after the Brexit vote; because interest rates were close to all-time lows. I expect to switch back when interest rates settle a little higher.
This is completely off-topic but I like to try solving the world’s problems on occasion. If the Fed Reserve were to raise interest rates (by manipulating fed fnd rates) to “normal” then this would mean additional debt servicing by $600 Billion, assuming “normal” is an increase of 3% interest rate. I’m not sure what “normal” really is.
So instead of spending an additional $600 Billion per year, risking deflation, blind mindless indiscriminate job losses, possibly driving us into a depression like occurred in the mid-late 1930’s, the next president or congress could actually consider some options, such as…
Paying 12 Million oldsters currently working $50K per year not to work. The precedent was already set when the government paid farmers not to farm. 12 million people is a staggering number. What does this achieve? It relieves some anxiety about the future of the elderly who think they can’t retire. It will put $12 Million youngsters to work, it will create labor demand, and it will generate consumer spending and inflation that has been missing up till now. It will reward the working class and wannabe working class as opposed to the elite.
An alternative to the above would be to use the $600 Billion to introduce an unequal bank savings rate. Say the first $500K of savings earns 6%, next $500K earns 4%, $10M in savings earns 0.1%, etc. This would encourage people without wealth to save without having their savings decimated by fed reserve policy. And it would encourage people with wealth to spend or invest their excesses.
Now obviously these ideas are very radical, but I do believe that the current massive debt is rendering the Fed Reserve useless. And interest rate manipulation that the fed does is not effective in a generational industrial revolution. They demonstrated that back in the 1930s (assembly line putting people out of work).
There - I’ve solved the world’s problems. Now its time to knock back another cold one
Steve
Going back to the subject under discussion (somehow I missed the initial posts), contrarian investing means investing contrary to prevailing sentiment. In other words, investing in stocks that few people are investing in while avoiding stocks that lots of people are investing in. Ditto with factors.
This is the cornerstone of my real-money strategy, and I’m up 24.5% YTD holding 15 to 20 stocks at a time (that’ll mean a CAGR of over 50% if 2016 continues along this route). My most heavily used factor is simply avgdailytot(90) with lower numbers being better (with some liquidity rules, of course). I also like investing in stocks with low turnover, i.e. vol/float.
The other key to being a contrarian quant investor is avoiding the most common strategies. I deliberately don’t use the TTM P/E ratio; ditto with EBITDA/EV. I deliberately pay no attention to ROE but use Greenblatt’s retun on capital instead. The accrual ratio I use rewards companies whose earnings are substantially below their operating cash flow–that’s a pretty contrarian position. I favor companies with a middling year-on-year sales growth–again contrarian. I look for stable operating margins, while most people like to see those grow. And price movements–past prices in general–play absolutely no part in my investment choices. That’s probably my most contrarian move.
Even when I apply these strategies to a smaller universe of more liquid stocks, they still work fine. I think contrarian investing still has a healthy future. There are always going to be market inefficiencies, and the whole job of a skilled investor is to exploit them.
