Search results

I was specifically asked by @truetype how I manage to perform SVD on very large matrices to get principal components. I gave a general answer because (a) I don’t know enough details as I am using someone else’s code and (b) this particular calculation is not particularly important for me any...

Transaction costs sensitivity

for annual data, you can check measuringworth; has cable since the beginning of time (1791): https://www.measuringworth.com/exchangepound/

You have a better numerical recipe you'd like to suggest, I take it?

In cross-sectional cases where there are relatively few assets and a lot of data, it's fairly easy to chunk it (many commercial/OS libraries do that for you). In longitudinal cases (time series to time series, that's where I get these large matrices) it naturally becomes a "special case" since I...

Are you a market maker in futures? Yeah, thought so :) It has nothing to do with "getting paid in bitcoin" as a speculator. A derivatives market maker is NOT as speculator, she/he does not like taking delta risk unless amply compensated for. In a world where there is no continuous liquidity in...

Futures were a failure because there is no real tie to the coin itself and no MM wants to warehouse that risk. As for the top, I honestly have no opinion. Is it going to zero from here? Possible, but unlikely. Is it going to 50k by the end of the year? Also possible and even more unlikely. If...

I rarely do that but nonetheless:

@Kevin Schmit Lol, I meant one million a side - that’s what you get by typing from the cell phone

I think it's 4x4 Xeon (Broadwell?), but I don't remember. It's blindingly fast for everything, including silly things like 1K x 1K SVD. PS. It's not mine, obviously - too grownup of a toy

Nope, it's a single box :) NinjaTrader, of course.. just kidding. I got a bunch of different backtesting engines, some are built for higher frequency stuff and deal with tick data, some are more geared toward minutely data for stuff that's less sensitive to latency. In reality, it's the...

Care to expand? There are plenty of non-MM players that have strategies across the time-frame spectrum as well as across a range of instruments. That does not mean they collect commissions.

To give you a sense, 1 Day of ES full book data is easily 1GB. You need about 2X of the dataset size. If you are testing across several weeks, it's pretty east to fill up a TB, especially if you are doing some manipulations and have a few different assets concurrently. It's freaky how cheap...

Well, I'd recon if Paul Newman wore a Timex it would get a pretty nice bid too

No, I did mean RAM :)

Tick or (or 1 min options) datasets get to be pretty large. At work, I have a nice little machine that has 1TB of RAM and I am probably ordering another one like that :) Previously, I would have gone the DIY route for sure. This time, I am not sure sure - it's shocking how cheap the refurbished...

Ok, explain this to me. Why would anyone want to pay anything for a watch that is, by construction, less accurate and durable than a quartz/electronic one? Also, if you paid X thousand for a thing like this, wouldn’t you be always scared to lose it or break it?

So, I have decided that I need to have a way to test some ideas while at home. Since there will be bunch of people chiming in, I will make it structured: Speclets: 128GB RAM 4TB+ hard drive Drive mirroring Headless (single VGA monitor sufficient) Linux Constraints: Price: it's for home so...

I just ran wc on my code for one of the more complex strategies - it’s about 6k lines :) I am pretty sure it can be done in half of that now

It does not have to be an all or or nothing reaction. It’s a scale ranging from doing nothing and allocating a meaningful portion of the GDP, I think the right answer lies with some sort of cost/benefit analysis. Obviously, that’s not what happened.