Let me try to be helpful.

If you have some kind of factor that you think predicts future stock returns (or similar) and you are making charts like below, then here are some tips...

We'll go through an example of trying to "time" SPX with the level of VIX.

A thread 👇👇👇👇
You get daily SPX index prices and daily VIX close data

You align them by date and plot them on dual axes, in true RealVision style.

"SPX tends to go down when VIX is high. I can therefore time an SPX allocation based on VIX. Let me share this on twitter" you say.

No. Very no.
There are two main problems with what you did:

1. The SPX price drifts. We can't directly compare the price of SPX in 2004 with its price in 2021

2. As traders, we are more interested in whether high VIX is *followed by* decreasing SPX prices, not *coincident with* them.
We'll deal with them in order.

First, you almost never want to be analyzing asset prices, because asset prices tend to be highly "non-stationary".

We can't compare them across our sample.

SPX 1000 in 2015 is not comparable to SPX 1000 in 2021 (which would be more significant)
Plotting prices is maybe how you're used to looking at stock charts, but it's no way to do data analysis.

If I presented something like this to my old boss he'd throw it back in my face and say:

"James, I need at least a modicum of f***ing stationarity!"
He'd then elaborate on my worth as a researcher (and human being) in ways that might appear problematic in the 2021 workplace.

But it was ultimately very helpful.

And, on Twitter, I hope you can benefit from this advice, even if you don't have a boss like that...
So how do we get him his "modicum of stationarity"?

An easy and effective way is to look at *differences in prices* (or returns) rather than prices themselves.

In this case, we're going to look at daily log returns.
We see that returns are much more "stable" and comparable over time.

10% in 2005 is comparable to 10% in 2021.

If we plot returns as a histogram, we get the familiar "bell-shaped" distribution, with a high peak, positive mean, negative skew, and fat tails.

Plot everything btw!
Now let's turn our mind to the VIX series...

VIX is much better behaved. VIX 20 is comparable in 2005 and 2021.

The histogram of VIX values shows it to be positively skewed (occasionally we observe very large values). This is also evident from the spikes in the chart.
So we'll leave VIX alone for now.

Now we ask the question

"Do SPX returns tend to be lower on days when the VIX closes at a high value?"

You can probably tell by the way I've phrased the question, that I think this question is a bit dumb, but let's soldier on...
We plot a scatterplot of SPX returns against the VIX level at the end of the day.

We see a weak negative relationship.

On days when the VIX closed high, we saw lower returns, on average.

Means we might be onto something, right?

Your job as a researcher is to call bullshit on everything.

"What, other than the thing I hope is true, could be causing this?"

Call yourself on your bullshit before someone else does.

But first, what else to you notice about that plot?
You should notice that the scatterplot is "noisier" and returns are more spread out for higher values of VIX.

This is because:
- we have fewer observations for very VIX
- by definition, the spread of returns is higher when volatility (VIX) is higher.
We don't have to do anything about this, other than understand it... but we can.

We can normalize our returns by their expected volatility by dividing our SPX returns by the previous day's VIX value

Now our scatterplot looks like this (we've controlled for that obvious effect)
So what's causing this - other than VIX having magic SPX direction predicting powers?

We're looking at the relationship between SPX returns and the VIX level *at the end of that period*.

We're looking at a coincident relationship.
If we know anything about equity index volatility, we know that volatility expectations will tend to increase when the SPX goes down.

So it's likely this is what we're seeing here.

Let's quantify it.
Here we plot SPX returns against the change in VIX in % points over the same period.

We see a very clear negative relationship. On days when SPX goes down, VIX tends to go up (and vice versa).
So, to what extent is this causing the effect we saw?

We can control for this by asking a better question:

"Do next day SPX returns tend to be lower, following a high VIX reading the previous day?"

We're now starting to deal with our second problem - the coincidence thing...
So we plot vol-normalized SPX returns against the value the VIX index value at *the previous day*.

This represents something we could *almost* do in the market.

(We can't in reality because the VIX daily marking was, until very recently, 15 minutes after the SPX close.)
Even though we've made some flattering assumptions about being able to see 15 minutes into the future, there's little evidence pointing at a significant relationship between the VIX level and next-day SPX returns.

This shouldn't be a surprise. The market is dominated by noise.
We're dealing with daily data here. A scatterplot of returns vs factor is always going to be a tough ask at this scale, even for the most venerable of anomalies
But I wanted to demonstrate that what *seems* like a clear predictive relationship if you're looking at it sloppily, starts to look like a mass of randomness when you analyze it more carefully.

- "Give me a modicum of f'ing stationarity" - prefer to work with returns rather than prices
- Define clearly the question you're asking of the data
- Plot everything
- Don't bullshit yourself... search hard for reasons outside of your pet theory

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Robot James

Robot James Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @therobotjames

9 Mar
On "stationarity"...

When we talk about something being "stationary" we mean that the observations look like they could be drawn from the same "bag of observations" (distribution), regardless of what time we choose to look at.

You can observe this by eyeballing charts.

VIX (blue) stays within a range the whole time. If it gets to extreme values, it's likely to revert back to moderate values.

By contrast, the SPX price (orange) just seems to drift away. It doesn't appear anchored to any range.

We can also see this by sampling from the distributions at different times.

Let's divide our sample roughly in half (2004-2012 vs 2013-2021).

From the histogram, it's clear that SPX prices are not drawn from the same distribution in the first and second periods

Read 6 tweets
28 Feb
This is a fantastic initiative from @KrisAbdelmessih.

It is hard for a beginner to know what effective trading "looks like", and where to concentrate efforts.

This will help bring the best of pro trading training programs to the masses.

Some thoughts on trader training... 1/n
The path to profitable trading is straight and narrow.

Having a theory on inflation, or memetics, or an in-depth knowledge of the VIX expiration, is a waste of time if I'm quoting corn options.

So good training concentrates the trader on things that matter.

You need to concentrate on micro things (information, motivations, constraints, game theory) when you're playing micro games.

And you need to concentrate on macro things (stats, broad behavioural tendencies) when you're playing macro games.

Read 12 tweets
10 Feb
Broadly, there are 3 types of systematic trading strategy that can "work".

In order of increasing turnover:
1. Risk premia harvesting
2. Economically-sensible, statistically-quantifiable slow-converging inefficiencies
3. Trading fast-converging supply/demand imbalances

1. Risk Premia Harvesting - is typically the domain of wealth management, but it's important to any trader who likes money.

A "risk premium" is the excess return you might expect over and above risk-free cashflows for taking on certain unattractive risks
"Equity Risk Premium", for example, is how you say "Stonks, they go up" if you work for Blackrock

(Though, they're really referring to the extent to which they "go up" more than an equivalent less risky thing)
Read 20 tweets
3 Dec 20
Trading is a game of repeatedly asking:

1. What exposures do I want?
2. What exposures do I have?
3. How do I get closer to 1 from 2, given that:
a) it costs to switch positions
b) my estimates are noisy
c) co-movements of assets are somewhat predictable

Concepts like "open trades" and "unrealized p&l" tend to be unhelpful in this paradigm.

If you don't like the exposures you have, then move them closer to the ones you want.

It makes no difference if you're underwater or in profit in your "position accounting"

There is no difference between a position that you have kept on the book for a while and one you just opened. It's exactly the same exposure either way.

Let's run through these 3 questions using a simple toy trading approach...

Read 21 tweets
2 Dec 20
In this brilliant article, Kris talks about the importance of trying to "shorten the feedback loop"?

How does a fundamental manager "shorten the feedback loop"?

I don't do anything that looks like "fundamental investing" but I couldn't stop thinking about the problem...

I think of the returns from "fundamental investing" coming from two sources...

1. Risk Premium - The tendency of risky assets to be relatively cheap vs their expected cashflows. This leads them to "carry" more than they would if their real cashflows were riskless

2. Mispricing - For behavioural/structural reasons, some assets are under/over-priced vs a reasonable estimate of their ex-ante "fair value".

On average, we expect them to converge towards fair value over some long time horizon.

This is "alpha" in fundamental investing

Read 10 tweets
2 Dec 20
Traders often implicitly assume recent conditions will persist - without checking whether that's likely to be true.

You need to understand the assumptions you are making, and whether they are reasonable.

Here are some examples and a simple analytical approach 👇👇👇

"I want to find an asset with characteristics X

"So I'm going to look for stuff which showed these characteristics in the recent past - and I'll hope it carries on having them for a while."

Sometimes this is reasonable. Sometimes it is wishful thinking.

How do we tell?

Here's an example of something we know is persistent - volatility.

"I want to find a low volatility stock - so I'm going to look for stocks which have been low vol this year"

Is this reasonable, or wishful thinking? Let's see...

3/n Image
Read 17 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!