Senin, 29 Juni 2020

An Update On Why Quantitative Funds Are Losing Coin Recently

A even out only came through Dow Jones newswire ("How Black Boxes Became Pandora's Boxes" past times Spencer Jakab) suggesting that recent losses are due to element models gone bad. Given my expressed distaste for such models, that should convey been my begin estimate instead of blaming the "exotic" models!

Minggu, 28 Juni 2020

A Reader's Comment On Quant Funds' Losses

A reader of mine (who want to rest anonymous) pointed out that virtually of the losses appear to come upwards from low-frequency trading models, piece high frequency models popular off along to perform superbly. This too confirms my ain experience. My enthusiasm for high frequency trading was expressed previously here as well as here.

Cnbc Interview

Folks, I volition endure appearing on CNBC tomorrow (Tue, August 14) for a alive interview virtually quantitative trading amongst Maria Bartiromo. The segment volition endure aired to a greater extent than or less 3:20 pm ET.

Sabtu, 27 Juni 2020

Cnbc Interview Video

Here is the link to my interview on why quantitative models are losing coin of late:

Jumat, 26 Juni 2020

The Robin Hood Regime

It has move apparent to me inward the final calendar month that in that location has been a massive transfer of wealth from the gigantic hedge funds running ingredient models to many day-traders amongst accounts less than $10M. I telephone phone this the Robin Hood authorities (regime being a mutual technical term referring to a detail trading environment, every bit inward "this is a mean-reverting regime"). Many, many day-traders that I heard from get got had i of their best months inward a long while. Is this exactly luck, or is in that location a deeper explanation?

I believe that in that location is a philosophical divergence betwixt ingredient models as well as many of the mean-reverting strategies that day-traders similar to employ, a divergence that plant to the day-traders' favor. I scream upwards a wise musing from i of my one-time bosses: he believes that a trading strategy volition survive profitable inward the long run exclusively if it performs a service for other marketplace participants. The service that mean-reverting strategies performs is the provision of liquidity, inward particular, short-term liquidity. What service does ingredient models provide? They appear to survive exactly arrogant bets on the correctness of the managers' convictions. For e.g. I believe that stocks amongst skilful earnings volition rising inward value. Or, I believe that stocks amongst increasing toll momentum volition hold inward that momentum. True, close of the fourth dimension the convictions of the best managers are correct, as well as many of these convictions are truly mean-reverting too (for e.g. the "value" factors). But on average, a ingredient model may convey away every bit much liquidity from the marketplace every bit it provides. And sooner or later, exactly about of these convictions are wrong. Maybe non incorrect for really long, but long plenty to displace investors' panic. This may survive business office of what nosotros are seeing recently.

Now am I advocating that every gigantic fund but exactly switch from ingredient models to pure mean-reverting strategies? No: that would survive impractical when the portfolios involved are inward the tens of billions. If everybody run mean-reverting strategies, in that location volition hardly survive whatever mean-reversion left to net from. (Look what happened to pair-trading inward the final few years.) When y'all are an investor inward a multi-billion fund, as well as y'all hold off the fund to deliver higher returns than the risk-free rate, y'all exactly get got to convey that high short-term returns volatilities volition survive business office of the bargain, exactly similar whatever long-term investments.

Kamis, 25 Juni 2020

The Failed Factors Convey Reverted

As I said inwards my CNBC interview, investors merely got to hold out patient amongst the component subdivision models. Sure enough, nosotros are seeing reports that the large drawdown suffered past times these models has already reverted every bit of Friday.

Rabu, 24 Juni 2020

Further Fighting On Element Models

A reader from a hedge fund (who wishes to rest anonymous) sends me about thoughtful comments almost element models. He has graciously allowed me to reprint them here:

"With regards to your weblog entry, 'The Robin Hood regime': this weekend I was truly too thinking almost the philosophy behind element models which y'all allude to inward the post. I am wondering if y'all own got whatever other thoughts equally to what service element models provide? Relegating them to 'just arrogant bets on the correctness of the managers' convictions' isn’t completely intellectually satisfying to me.

I facial expression at factors equally such: the returns I acquire for exposure to diverse factors tin hand notice come upward either because the marketplace is inefficient as well as systematically misprices those factors (alpha), and/or because I am providing about service via the exposure (and collecting about variety of guide chances premium associated amongst that service). My query #1 to y'all is, are y'all convinced that all of the returns to element models are indeed exactly from guide chances premiums as well as non alpha? If alpha exists, it’s less clear that a service needs to live on provided to the market, at to the lowest degree to me.

However, let’s assume (as I believe your boss did) that inward the long run, the marketplace is efficient. Then, y'all volition live on compensated for element exposure entirely past times bearing about guide chances or providing about service. In my mind, about detail conviction of a managing director doesn’t necessarily qualify for a guide chances element inward as well as of itself - I recall nosotros concur on that point. But are at that topographic point possible fundamental, valuation-based explanations behind these factors? Perhaps depression VALUE companies are mostly those companies amongst bad recent surgical procedure but which are expected to turnaround / mean-revert (as y'all somewhat propose inward your post) as well as the guide chances y'all conduct when buying a depression P/E fellowship is “turnaround risk”. Or mayhap high MOMENTUM companies are companies riding an manufacture tendency as well as y'all are bearing “trend continuation risk”. So, my query #2 to y'all is, are y'all convinced that at that topographic point are no such explanations?

If element models produce indeed work, it seems to me that at that topographic point must either live on existent risks behind the factors, or alpha, or both."

And hither is my response:

"I believe the service that about value factors render is the efficient resources allotment of uppercase to those companies that deserve them, exactly similar whatever value investors do. In this case, the factors promise to position these companies faster than humans can, as well as hence select uppercase to them sooner. I own got no declaration amongst these factors equally they too render liquidity, albeit on a longer time-scale. However, amongst regard to diverse momentum factors, they are inward fact exactly betting on for certain behavioral characteristics of investors, or on the ho-hum dissemination of news, etc. You tin hand notice debate that they render a service past times improving the efficiency amongst which data almost companies disseminate. But the work is that i time everybody are using these momentum factors, the marketplace becomes efficient as well as whatever farther bets generate losses.

So I am quite willing to select that many of these (momentum) factors stand upward for alpha, but these factors are generating to a greater extent than losses equally to a greater extent than investors employ them. I am too willing to select that many of the (value) factors stand upward for guide chances premia. As to a greater extent than investors employ these, the net goes to zero, but fortunately non negative equally the guide chances too disappears."