Trading algorithmically a folio without stops (with IB), real $$$

[fullautotrading]

It's some time we do not look at an order cloud. Here is one of the nicest (one layer with UPRO) [shot taken right now, while market is closed (some ticks are actually arriving for this instrument even now)]:

The picture at the bottom is exactly the same, but in "player view" mode. (This covers a span of almost 80 solar days, and 228 fills on this layer.)

Useless to say, that when we trade we need to always use the second mode ("player view"), which allows us to clearly see what's going on in terms of "open players" (closed players can be ignored, of course) and also to manually "inject" new strategic players, when we wish so.

 

[jb514]

I understand where you're coming from but I feel like you are neglecting single names. Try adding something as benign as YHOO or WMT just to get some uncorrelated moves in your portfolio.

 

[fullautotrading]

I understand where you're coming from but I feel like you are neglecting single names. Try adding something as benign as YHOO or WMT just to get some uncorrelated moves in your portfolio.

Thank you jb514,

I am adding them right now to the folio. We will keep them under watch for a few days, and will see and discuss here how they move (spread, volume, volatility, decay (if any), history, shape of price curve, amplitude of fluctuations, etc...). Then, if we like them, will "activate" them for actual scalping with some appropriate scalping/hedging "game".

Of course, let me know any other instrument we may want to take a look at and test.

 

[fullautotrading]

End of the week.
Today we had some move at the start of the trading day, which allowed to close (not all but) several players of the metals (PNL touched 21.8K at its peak). ERX scalped a little bit up and down but essentially remained where we left it. ZSL is a pretty poor "performer" lately (excessive spread, too little fluctuations) and I will throw it away as soon as possible.

The situation looks fine. We are proceeding in a very patient but methodical manner, feeding inexorably our statistical drift ((G-L)/L) [over 15%] and accumulating about 98K of "G-L". Over 6.6K gone in commissions for about 7K fills. No technical problems so far. Margin usage this week has been mostly in the range 140K-170K (150K right now).

Next week we should begin to take a look at the unleveraged instruments suggested by you guys (jb514 suggested YHOO and WMT). (Probably, to trade those we need to adjust a bit the scalping sizes and entry distances in the scalping/hedging game, and it's something which can probably also be done dynamically, while watching the instruments move. (Clearly, the best instruments will be those rich of frequent fluctuations up and down (of at least a few% points), which allow to open and close the scalps frequently). Monday, I will also upload a revised version, with various refinements I have been making during the week.

 

[AItrader]

thank you Fullautotrading for your continuous update of your trading activity

As you are planning to move trading leveraged instruments toward not leveraged ones, and maybe also from index/sector tracking ones to single stocks,
it would be interesting to keep track and talk about the adjustments in G-bot game parameters that you need to do, if any, in order to successfully moving from one type of assets to the other.

Keep up posted!

Best,

 

[fullautotrading]

thank you Fullautotrading for your continuous update of your trading activity

As you are planning to move trading leveraged instruments toward not leveraged ones, and maybe also from index/sector tracking ones to single stocks,
it would be interesting to keep track and talk about the adjustments in G-bot game parameters that you need to do, if any, in order to successfully moving from one type of assets to the other.

Keep up posted!

Best,

Hi AItrader,

sure. I am actually a bit "skeptical" about unleveraged stuff, because of the possibly "smoother" price waves, and the relative weight of commissions [actually commissions look a bit excessive in the current setting and capital, where the order sizes are relatively small and the number of trades still in the "max commission price" zone. However, those trades are still "necessary" to avoid larger DD, which would be worse than commissions]. Anyway, we can certainly give them a careful try (maybe start with smaller sizes).

Mostly, I have been considering so far futures and ETFs, and it actually turns out that the instruments actually "tradable" (for various reasons, like spread, leverage, decay, microstructure, general dynamic, ...) are less than one would expect. So, new good opportunities are welcome.

 

[fullautotrading]

Start of new week. Not much activity. Some unrealized pulling back and driving the PNL higher.

Also watching the two new stock symbols. Did not move much (remember that the green horizontal lines represent (about) 1% moves, compare with previous ETFs' charts too to get a sense).

 

[Trader13]

I feel like you should be trading less correlated assets. There's no reason to be trading both TNA, TZA or VXX, UVXY. Looking through most of the symbols on the bottom it appears as if most of what you are trading moves in lockstep with one another, most of which can be accomplished by just trading SPY.

Why don't you start slowly adding equities with low market correlation?

Fair point, but keep in mind that two time series with a zero correlation still move in the same direction HALF the time. See the attached chart for an illustration using two sine waves with zero correlation.

 

[fullautotrading]

End of the week. Some of the new profits have been reinvested in new moves. Drift is proceeding ok. Everything under control.
Most active instruments this week have been the small caps (TZA, TNA). We still need some activity in metals and energy to see some progresses in that area. Most disappointing have been JDST (always not shortable) and ZSL (large spread, no moves.) I will definitely remove them as soon as possible.

Here is the charts (a week) of the couple stks we have start watching (WMT, YHOO):

WMT moved very little. YHOO made a marked gap between sideways moves. (green lines represent about 1% moves in each chart) Not very exciting scenarios to trade with a scalping action, so far.
I have put under watch also a few other stocks too, to see how they move and in case activate them.

 

[fullautotrading]

Start of new week. Most of the "load" in small caps of the previous week has been released today, pushing the PNL beyond <b>27.6K</b>, at its peak. Margin usage is at the moment pretty low, just 128K. Metals and energy still stuck. Today I have slightly changed the settings, making the SDXs a bit more sensitive to "lock up" more quickly profits.
I have been working on ETFs classification, and corrected the categorization of some of the instruments (mostly metals) in the various games and position constraints. This should improve future performances. (This is a journey where we do learn/realize something new every day.)

I am now also looking a few stocks (amzn, bac, csco, ebay, pfe, x, yelp, wmt, yhoo, ... make suggestions!) and will soon decide which one turn on in "auto" mode. The general impression is the dynamic and microstructure is pretty different from ETFs. Anyway, will keep watching them for a few other days, and maybe looking for more candidates.