What objective method would you choose to identify volatility? Beta, ATR, ect ...? What time frame? Or would you use something more general like VIX?
Please Critique (Poke Holes) in My Entry/Exit Strategy
- Thread starter baggs
- Start date
What objective method would you choose to identify volatility? Beta, ATR, ect ...? What time frame? Or would you use something more general like VIX?
Ugh, as I mentioned earlier, I have a lot of experience trading (6 years). I understand that psychology is more important than all else. That's a given. That's also why I'm trying to go 100% non-discretionary.
This is the "actively managed" portion of my portfolio. The rest is in passive lower "risk" assets.
LOL... ok... so if you have lots of experience, and if you've been trading for years, you expect that someone will take the time to very carefully find the small flaws in your approach and help you extract every last percent from the market? You never shared any math by the way, so its not like there is even anything for anyone to go over.
If you are as experienced as you say, then you have all you need. Asking strangers here will only dumb you down. And honestly, if you are as experienced as you say you are, then you should already have the answers to your questions.
When it comes to trading, everything works, depending on how you do it. Trend trading and counter-trend trading both work. Scaling into losing positions can also work, for those who know how to do it. Everything works, but its all in the details. So for you to come here and ask a general question about scaling, given you have so much experience, seems kind of odd.
No, I would not look at VIX or some other derivative. I would look at the stock's price over time and calculate the relative daily return. Thus not the stock price, but the change compared to the previous day. Doing this over a longer period (e.g. 25 days) and calculate the average % change per day. This gives a good idea which stocks are less volatile and which ones are more volatile. You may want to experiment a bit with the lookback period to find a suitable solution for your system.
I think the important thing why he bought those stocks the first place.
I didn't catch if you mentioned it.. Is this a long only strat?
Not important. Perhaps I should have been more clear:
Does your system generate short trades? If so, I would use those and use 2:1 leverage on your longs. A semi-famous long/short portfolio manager (I forgot his name now) once said: over the 20 years I have been managing my fund, I broke even on the short side, but it has allowed me to make bigger bets on the long side (He was between 100 and 200% long and between 50 and 100% short)
These comments are more general, but can be applied to this problem:
- A trading strategy typically consists of the following items:
Alpha Model: this model defines the edge of a trade; it can have input information that includes market data, signals, etc. the output should be a theoretical price for the asset you are trading. We expect the asset price to converge towards the theoretical price.
Risk Model: this model dictates trade size/inventory management. For example, if an asset's current price is $1 and our theoretical price is $2, we might buy one unit with the expectation of making ($2-$1)=$1. However, if the price drops to $0.90, we now are faced with the following dilemma: if we are 100% confident in our theoretical price of $2, we now see $2-$0.90 = $1.10 in edge. However, we are currently down $0.10. The risk model should take this input and determine whether we should put on another unit of position, or possibly sell out some risk.
Execution Model: this model determines how an opportunity to add or remove inventory is actually executed in the market. It should account for execution cost (spread crossing/slippage) as well as market impact (less important for small order sizes relative to net liquidity).
As this relates to your question:
- It seems there is little separation between risk management and alpha. First determine what your alpha is, then worry about capturing as much of it as possible, while respecting risk limits. The question of when to scale out of position is really a question of whether there is still alpha in the trade. Here's an example:
Time 0: You see an asset at $1 and have a theo price of $2. You buy the asset at $1.
Time 1: The asset has moved to $1.50 and your alpha model predicts a theo price of $1.75. That is, you have made $0.50 of PL (unrealized) and have $0.25 of edge left in the trade.
Time 2: The asset has moved further to $1.60 but theo has moved to $1.60 as well. You have PL of $0.60 and remaining edge of $0. Therefore, you would sell out this inventory since it no longer has edge.
This might be a different way of looking at trading, but it is far closer to a professional type of management. Think about assets and theoretical prices, not dollar amounts because that is one level more abstract than the asset price. I think the most common example of this idea is a moving average strategy (not saying there is any alpha here, but this is the idea behind it). If one thinks that price will converge to some MA, then if price is below MA we would buy and if above we sell. This is simply a bet of convergence to theo, which in this example is the MA.
- A trading strategy typically consists of the following items:
Alpha Model: this model defines the edge of a trade; it can have input information that includes market data, signals, etc. the output should be a theoretical price for the asset you are trading. We expect the asset price to converge towards the theoretical price.
Risk Model: this model dictates trade size/inventory management. For example, if an asset's current price is $1 and our theoretical price is $2, we might buy one unit with the expectation of making ($2-$1)=$1. However, if the price drops to $0.90, we now are faced with the following dilemma: if we are 100% confident in our theoretical price of $2, we now see $2-$0.90 = $1.10 in edge. However, we are currently down $0.10. The risk model should take this input and determine whether we should put on another unit of position, or possibly sell out some risk.
Execution Model: this model determines how an opportunity to add or remove inventory is actually executed in the market. It should account for execution cost (spread crossing/slippage) as well as market impact (less important for small order sizes relative to net liquidity).
As this relates to your question:
- It seems there is little separation between risk management and alpha. First determine what your alpha is, then worry about capturing as much of it as possible, while respecting risk limits. The question of when to scale out of position is really a question of whether there is still alpha in the trade. Here's an example:
Time 0: You see an asset at $1 and have a theo price of $2. You buy the asset at $1.
Time 1: The asset has moved to $1.50 and your alpha model predicts a theo price of $1.75. That is, you have made $0.50 of PL (unrealized) and have $0.25 of edge left in the trade.
Time 2: The asset has moved further to $1.60 but theo has moved to $1.60 as well. You have PL of $0.60 and remaining edge of $0. Therefore, you would sell out this inventory since it no longer has edge.
This might be a different way of looking at trading, but it is far closer to a professional type of management. Think about assets and theoretical prices, not dollar amounts because that is one level more abstract than the asset price. I think the most common example of this idea is a moving average strategy (not saying there is any alpha here, but this is the idea behind it). If one thinks that price will converge to some MA, then if price is below MA we would buy and if above we sell. This is simply a bet of convergence to theo, which in this example is the MA.
With a long-time data, you have shown annual compounded 15% and current account has 900K.
With a simple formula 1.15^10 = 4.045558, your asset was roughly 225K (quarter of current asset).
And, based on the simple compounding formula, asset after 10 years is expected to be 4*900K=3.6M
(Likewise 3.6*4=14.4M after 20 years and 3.6*4^2 = 57.6M after 30 years)
1) I wonder how many more years you will be alive, hoping how much your asset will grow.
2) For the last 10 or 20 years, how much was your accumulated tax and brokerage sum?
Is the above annual compounding 15% the after-expense (tax+brokerage) or before-expense?