Banker bot in python

The actual snipppet I am writing.  Basically its a finite precision, windowing Huffman encoder, works over any python numerical, iterable sequence.  It creates a generator graph, based upon the submitted sequence.  Once the generator is created, we can feed it a uniform random sequence, and it will generate a pseudo random sequence of typical numbers.

We can use this as a portfolio management tester. Feed a simulated sequence of trades to your portfolio software, your portfolio software can compete against the bot.

From Wiki:

Adaptive Huffman coding (also called Dynamic Huffman coding) is an adaptive coding technique based on Huffman coding. It permits building the code as the symbols are being transmitted, having no initial knowledge of source distribution, that allows one-pass encoding and adaptation to changing conditions in data.

This is new stuff, I think.  Just remember. A compressed encoding for us really means typical sequence generator.  If we removed redundancy by quantizing to a fixed precision and re-denominating the trades, essentially. Think of as this, optimally grouping trades in the range of 1200 - 2450 into a new denomination, the mid-trade trade unit. A quant, which really represents a large chunk of day traders.  Stuff like that comes out automatically, day traders own some nodes on the generator graph.

My approach is to start with a simple basic Huffman encoder.  Then go search the literature for adaptions, like the one above.  My start will be based on a simple fixed window encoeder, and I will use whatever built in sort is required. Most of the sort is local, and down he graph, so it is not always a 'sort the whole thing'.  I will make that efficient. I have an advantage, I have a simple debugger and sophisticated trap machine.  From here I am really good at reading math papers and making code  of them.  I can make sure we get our Smart Card, on time, before the helicopters fly.

<My bot is non profit from the owners perspective, it will measure the precision of the market, makes its own quantization match. Thus transaction rates are adapted  and user sees lower transaction fees.

Understanding this

Imagine that paper money was smart in the following sense.  Ity monitored your purchases, and then changed he bill denominations so you typically pull out one bill for each trip to the store.  Over short periods, you your 'ATM' would start putting out unique denominations of paper, one bill might be at $5.50, because he bot it notices you have tacos at the truck every day.