pierrealexandre

There is some interesting research regarding the optimal tick size (by instrument), eg https://arxiv.org/abs/1207.6325

No, it does not sit at the same 1-tick spread all day. Typically for these tickers the 1-tick spread transitions very quickly to the 1-tick spread above or below (each individual market is going to be 2-tick wide for a very small amount of time at each tick transition, much less than 1 second).

If I understand Table 1 correctly, it shows that with ICC a block size of 8 with 4000x4000 matrices gives roughly a speedup factor of 2 (~10.53 / 6.38). The unit is not clearly stated but it looks like a bigger number means a better "performance" ie a faster program.

The article provides historical context and a framework to think about the question. I think it does a better job answering the question than your last 2 paragraphs.

Actually, replace 90 by 91 (100 / 1.1), and 35 by 38.5 (100/1.1^10)

You don't "specify a duration" for the valuation: an infinite duration is baked in the valuation itself through the time value of money. If the annual interest rate is 10%, 100$ in 1 year is worth 90$ today, 100$ in 10 years is worth about 35$ today and so on. Of course, in the case of company valuations, you have to also account for the uncertainty (both on the upside and downside) of the future cash flows.

Isn't the phenomenon you describe extreme price elasticity rather than inelasticity?

Also, any source for your assertion about the effect of rent control on rents in NYC? I'm very interested in this if you have it.

Are you aware of the distionction between market orders and limit orders?

HFTs by themselves can provide liquidity (like any other investor), namely by entering a limit order in the order book. They can also by themselves narrow the bid-ask spread, namely by entering a limit order at a price more competitive than the current first limit.

Of course, someone else has to send a market order to the exchange for a trade to execute.