Well, we had a great session on Thursday, but I know many of you missed it. The new software I was using (GotoWebinar) did not record the session, so unfortunately there is no video replay. Instead, I will provide a summary of the topics that we covered. This will be nice so we'll have an archive of the overall topics.
Topic 1: Beta Weighting Portfolios
When analyzing the overall risk in your portfolio, one of the biggest things to watch for is how your portfolio is expected to perform relative to the market. That way you know what daily p/l you should be expecting so you know when to adjust/cut positions.
So the technical term for beta is a "number describing the relation of its returns with that of the financial market as a whole." It's related to modern portfolio theory and CAPM, but I really don't care about that-- we can simplify it for our purposes.
Essentially, the more volatile a stock or etf is relative to "something," the higher beta we have. So if the market moves up, we should expect our higher beta names to move up faster, and vice versa for lower beta names. Examples of higher beta names are small caps and anything with china, solar, and green in the name. Lower beta names include utilities and names like WMT MSFT etc.
A beta of 1 means that it is perfectly correlated to the market. A beta of 0 means that it non-correlated to the market.
A beta of 2 means that for every point the underlying market goes up, the asset will make 2 points.
And that's beta for a single name. But beta works in aggregate-- we can sum up the beta in our entire portfolio.
If your portfolio has a beta of 300 SPY, it's equivalent of owning 300 shares of SPY, and you should expect the equivalent volatility (and a bit less if you're a good stock picker).
Calculating beta is beyond our scope, and it's related to variance and some other calculations. But thinkorswim and other brokerages can do it for you, and that's what we look at when I do our portfolio review.
The first calculation you want to make is the max tolerable beta that you should have in your portfolio. So first get the current price of SPY-- 114.57. Next, take your overall portfolio size-- we'll do 100,000 to make it simple.
Now divide SPY into portfolio size and we get 872.8 -- that means we can at the current price get exposure to 872 shares of SPY. That is the maximum amount of beta (variance) you want-- unless you want to be leveraged up.
Notice that in our calculation that as the SPY goes up your max beta exposure goes down-- but that's under the assumption that your portfolio size remained constant. Ideally your portfolio will increase because you're a good trader, so your max beta can increase.
The next calculation to look at is market variance vs. beta. That calculation will show you what your daily "risk" should look like.
Let's say that our current beta is 500 SPY. And the current average true range (ATR) is 2. So you should expect your p/l range between +/- $1000. From there you can see how you do day to day-- if the market is down 2 and you're only down $200, your portfolio is doing very well. If the market is flat but you're down $2k, there's a problem and you need to reassess the risk that you are currently exposed to.
I use ATR, but you can use other volatility measurements-- derive movement based on current VIX readings, BB Width, etc. ATR is the simplest for me.
Topic 2: Mean Reversion
"Mean reversion" is the term many people use when a position is going against them. It's similar to "averaging down" and "value investing."
For us, it is a shortcut to understand how stock prices are modeled. When it comes to options and how they're priced. Stocks resemble geometric brownian motion. That's a stochastic process that means price is lognormally distributed.
All that means is that the change from day to day is going to resemble a bell curve-- that makes sense, because we see on average a 1% move in the markets. That "average" move depends on the volatility.
So mean reversion in essense is that the change of a stock has a certain mean, and when we see a stock move 4% and the average change is 1%, we don't expect the stock to continue moving 4% per day-- and we can construct option plays to take advantage of that.
Mean reversion works only a certain amount of time. And then we have LTCM. And 2008. And LEH, BSC, and so on. And you can make money 8/10 times on the basis of mean reversion, but those 2 times can wipe out your gains if you don't manage your risk.
Topic 3: Historical Volatility vs. Implied Volatility