Winter time - stress test for traffic congestion and car condition

During winter I witnessed many traffic jam or just slow traffic problems. Driving takes sometimes 4 times longer than in average conditions. Sometimes slowing down below some level in just few critical points is enough to clog up the roads in the whole city. I see some similarities with IT systems during stress load.

Making new roads is not real life solution for "fix it now" traffic situations. More appropriate is holistic analysis of the system performance and system tuning, like smart adjusting traffic lights to actual congestion etc. Some car drivers are responsible and try to help everybody moving in. But most of cars are moving like flowing water stream, filling every empty space in shortest time. So kind of central traffic steering makes sense, and is subject of scientific research and real life implementations.

Another situation. I haven't problems with my car during winter till -10 centigrades. But few more freezing days made little inefficiencies in car systems stack up and disabled my car for few days. Something like total performance degradation in IT system while stress test.

It's good to have backup plan for more harsh "freezy" days.

My first Mechanical Trading System results

As you may already know, my "new hobby" are Mechanical Trading Systems. After first quarter of executing transactions generated by my first Mechanical Trading System I have gained about 0% (literally almost ZERO). It may be surprising for somebody but it looks not so bad (but could be better).

Here is why:

• There were transactions costs, low enough for about 40 transactions, but still were.
• There were execution price slippages - sometimes it's hard to execute order at given price
• Some initial errors in systems gave me some loses too (about 10% of initial capital).
• I've made cross-validation procedure, but testing data set was too small to get good statistical meaning.
• Tested gain values were low (but still positive) and comparable to values without losses caused by "external" errors.

Statistics showed that the system was performing better in training period, and worse in testing and then in real account execution period. Underlying model was extremely simple and had above 100 transactions per year to avoid perfect fitting to one or few best price movements. But that is not excluding potential over-fitting on some kind of higher level longer horizon component of price movements, like up/down trend during sub-periods.

Results are definitely worse than gains from risk free return at a bank account. But it was relatively low cost exercise and I have learnt something new about practical details of investing, trading systems and statistical validation. Maybe next approach would be better.

So, I'm doing research on new simple model with higher gains/lower drown downs in test set. This time I have to make better validation tests to be more sure that system has generalization capability, and in future it could keep assumed risk/gain. That means more testing data and implementing better search procedure.

Calendar reminds me of plans for the future

End of year is good moment to rethink what happened through past months, and redefine some goals for the future.

Christmas and weekend days off, make good opportunity to take some time to think about it (besides family time and partying with friends).

I'm finishing older projects so there will be time share for new ideas. I'm doing some research and trials in Trading Systems, and I'm going to continue that. There is some place for personal development too.

I will be looking also for ideas for something more sustainable like startup project that could survive. It should be interesting. And fun.

Reviewing printed vs on screen documents

I have observed interesting phenomena: reviewing printed document reveals more mistakes in text than looking on the same document on computer screen. And I think I'm not alone - I've seen other people had same problem too.

It has to be something with focus on a task of document reviewing and ability to concentration in given context. Computer screen vs paper document. What is this all about? Is paper document better to read? I think it's related with casual context related with each of the media.

My computer (screen) context is something like:

• work (reading, writing, drawing, programming),
• news,
• learning,
• entertainment (movies, games),
• sharing some functions with TV,
• promotes multitasking,
• contains more chaotic and less quality content that needs more filtering and preprocessing

Paper context:

• books (facts and fiction),
• documents (more important and less like just printed electronic versions for convenience),
• better quality content that needs focus.

Time spent at each media is about 95% for computer screen and 5% for paper, so rare paper usage can make it potentially more interesting for brain too.

While using computer I'm doing many things while multitasking, and filtering low quality information that is reducing deep focus. Paper media in comparison switches my consciousness into context that makes me more deeply focused, less distracted (no multitasking), and and that makes me finding fine text errors easy. I don't even need to think about it, it's how my brain programmed itself in choosing proper context. It's not strange that computers (and Internet) turn me into "information trash" mode when I can't see details.

I'm not mentioning eye-strain effect, because of good modern LCD screens and readers using electronic paper so impact is small and difference between media in that area is vanishing.

Idea, how to stay focused on text without paper version of document, is to read text in simple non distracting computer environment like full screen mode, no other windows, bells and whistles (easy), and treating it like high quality content with meaning (that's harder). It may need some time to rewire brain, but probably that's the way to go.

Genetic Programming for finding trading system algorithm

I've been thinking about using Genetic Programming (GP) approach to find new trading system algorithm. There is no simple relation between such algorithm and outcome of the system (profit/loss benchmarks on historical data). To evaluate usefulness of algorithm, it have to be tested on real data. So its good target for that kind of automation.

I have been using Genetic Algorithms (GA) and even built some kind of GA framework at the university. Now I'm browsing through resources to refresh basics and find out what is new in the subject. I'm reviewing also GP frameworks to use (choosing possibly the simplest one). Results could be interesting - but it needs some integration work to take off.