Trading Model Terms

Sum of Open and Closed Profits for test sample

Current Closed Profit for test sample

Current number of share long or short

Total number of trades the model made on the test symbol during the test period

The average gain for all trades the model made on the test symbol and test period

The average number of bars a trade is held

Average winning trade

Average losing trade

Maximum winning trade in the test sample

Maximum losing trade in the test sample

The average number of bars a trade is held

Average winning trade

Average losing trade

Maximum winning trade in the test sample

Maximum losing trade in the test sample

The Maximum consecutive wins in a row

The Maximum consecutive losses in a row

The percentage of winning trades

The ratio of profits to losses

profits
---------
losses

The largest closed dollar draw down the model experienced in the test sample

The largest open dollar draw down the model experienced in the test sample

Ratio of Closed Profit/Max Draw Down

Closed Profit
------------------
Max Draw Down

Ratio of Open Profit/Max Draw Down

Open Profit
------------------
Max Draw Down

How many calendar days it took to reach a new CLOSED equity high from the previous highest CLOSED equity high

How many calendar days it took to reach a new OPEN equity high from the previous highest OPEN equity high

A measure of model robustness in a test. This ratio answers "how many of the winning trades can be removed for the test to become equity neutral

An evaluation model for testing a system .
As its name implies, it pessimistically weights profit (by reducing it) and loss (by increasing it).

How many bars will the model wait before adding back to a position after achieving the first profit level

The amount of shares the model can purchase to risk on each trade

After the first profit objective is reached, additional profit targets are laddered in the market according to the original profit level.

PROM adjusts the gross profit by calculating a new, pessimistic, lower gross profit. The first step is to calculate the number of winning trades reduced by its square root or adjusted by its standard error. This adjusted number of winning trades is then multiplied by the average winning trade to arrive at a new, lower gross profit.

When we do prom sorts, okay; which were still a good sort, okay, what you want to do is you want to have that outer sample as big as you possibly can, okay; so that you’ve got - you need to get at least 16 or even preferably 30 trades in that outer sample, okay. 16 minimum; anything less than 10 or etc. forget about it; you’re not evaluating it strong enough, okay.

RTN is listed in decimals, so a Number of .10 would mean you could remove the top 10% results of this parameter set and still break even. Any model that has A RTN above 5 is good. Remember To multiple RTN by the TCnt to come up with how many trades RTN Actually represents. Typically, you can increase RTN by reducing trade Frequency.

First of all, you had a trade count of 78, right. 61 percent of those were winners, okay. So, how many winning trades did you have? You had 78 times 0.62; that meant you had 48 winners, okay. Now, what return to neutral does is that it takes those 48 winners and it organizes them from the highest winner to the lowest winner, okay. And then it says, if I remove starting from the highest, I’ve got them removed 20 – how many of them do I have to remove to get me to break even. So, you would have had to remove 23 percent; so times 0.23; so you would have had to remove the 18 best winners and you’d have been at neutral.

For example, if we buy something at $20 And we risk $1, what is 75% of our risk? .75. So, at 20.75 will sell 80% (or some Other amount of our position size) and have The remaining 25 shares (20% of position) Follow the market using a trailing stop.

Rough estimate – whatever your iterations are you need to test on 10 times that as a sample.
Degrees of Freedom are your rules plus your conditions, your iterations, your steps, or every place you have an X.

you take the observation the event (trade count) and find the square root of that number. It the TC optimizer model, this number is express as a percentage.
Basically, it says the higher the number of observations, the greater the confidence you can have in a model. For example:
The sq root of 10 trades is 3.16 or + or - 31%
The sq root of 50 trades is + or - 14%
The sq root of 300 is + or – 4%

Rough estimate – whatever your iterations are you need to test on 10 times that as a sample.
Degrees of Freedom are your rules plus your conditions, your iterations, your steps, or every place you have an X.

For example, in MQModelx 1st parameter is left peak, and if we have 15 possible combination's of parameter “left peak” degrees of freedom are 1 (left peak) + 15 (condition) = 16
If the 2nd parameter is “right peak”, we would have 1 (right peak) + 15 conditions = 16.
Next, we have another rule called offset that is 1 + 0, 6 * 8=48

Is your first profit objective in relation to your risk. For 100, you will take off 100% of your risk at your objective. A limit size of 80 means you’re going to take 80% of your position off at the first price target. Optimizing on the limit size means answering: what is the most profitable size to sell (buy) to protect your profits.

This parameter measures risk to profit objective. In optimizing for 75%, 100%, 125% or 150% of our risk, we are testing For profit in relation to risk we should Set our profit objective given the past data.

MAE assesses each trade from the strategy and determines how much a trade had to lose in percentage terms before being closed out for a winner or loser. You put on a trade and if you are like most traders, the position will move against you.