KVB do you paper trade your models first?

[jjgold]

I want to know

 

[BobbyFK]

After seeing those videos I'm sure he does. He looks very detail-oriented when it comes to that stuff

 

[flyingillini]

 

[jjgold]

I’m gonna be interested in his answer

 

[BobbyFK]

I’m gonna be interested in his answer

@KVB

 

[KVB]

I’m gonna be interested in his answer

The answer is yes.

Ok, so first with some basics.

Traditionally, and normally, when you have a data set to build some kind of logical regression model, or even to train one, you divide the data set into training and testing sets.

I might take 2/3 to 3/4 of the data set to build, train, and mold into an exquisite and fine tuned delectable product to take to market and compare, but how do I know it's still predictive?

I don't, in fact, I never really do for sure, it's a matter of probability.

What I then do with my scrumptious piece of juicy inputs to outputs is to put it to the test with the remaining 1/3 to 1/4 of the data.

That, my friend, while being historical, is in fact paper trading.

If it passes that test, against results and market prices, to produce an acceptable margin and often closing line value, which translates to profit, we go live to market with money with the amount often depending on the strategy and it's predicted predictive performance.

But I have mulitple approaches that can fall in and out of favor with the markets. Those, after having been through the process above, continue to be tweaked and tested against the current markets. That, too, is paper trading.

So yes, models are paper traded but I don't get to that point without first performing all kinds of tweakes and tests.

Paper trading is important as the market landscape is in a constant state of change, like the shifting sand dunes of Ngorongoro, Tanzania, or even the Namibian coast.

If you enter that desert without a gameplan, goal, or confident methodology, then you are sure to get confused when the landscape changes around you and that confusion will lead you to abandon what gameplan you had, or break discipline, allowing variance to get the best of you.

Don't go to Namibia or Tanzania on a windy day without a map or compass, @jjgold ; and, likewise, don't spend your hard earned money on untested sportsbetting trends without some degree of scientific confidence.

Short answer, it's all paper trading until it's not.

But once it's past the paper trading stage and we are going to market with confidence then...

Good Luck, Gold.

 

[jjgold]

What an answer unbelievable