Building a systematic system part one - data capture

let's say your trading strategy is based purely on Earnings surprise or maybe actual Quarterly earnings change, both a fundamental data, the announcement will usually happen after the close. If the info was pushed through a "pipe" and your trading app capturing it, it couldn't act on it because the market would already be closed. A simple pull at the end of the day for "earnings" change could update your metrics table for the next morning trade.

let's say you missed the earnings news, would it be a tragedy ? if you are really long term fundamental, your trading strategy would rely on a number of factors, therefore a single data item change would not be enough to trigger a trade or change dramatically your portfolio structure. Further more, even if you caught the earnings news on time, it's more than likely than the market would have reacted already, hence making your trading decision pointless. It's also more likely than the market reaction will reverse soon in a few days after the over-reaction to the news is being "digested". I see this happens times and times again, like clockwork.

Either way, it shouldn't impact your trading strategy long term view of the securities you are holding for that strategy since other factors and a periodic cycle would only come into play for your real decision making process.
 
it's important to remember than Fund managers are not traders, and you should always have a long term view of your strategy

Hedge Funds are more like traders, hence their erratic returns and high failure rates.

short term strategies are more likely arbitrage strategies, that means they are on the clock and they tend to disappear fast, usually with a disaster looming at the end if you put too much trust into them.

Short term strategies can't simply rely on long term fundamental data because the data cycle is simply too long (duh!), they can only exist if they "react" quickly to certain data, and usually these are data with short cycle (like prices and volume). The whole HFT industry is like this, it's a game of ping-pong with random chances for 90% of the HFT players out there.
 
Butterfly...can you at least once try to be a little reasonable? You were the one who brought up multi billion dollar funds. No billion dollar fund trades fully systematized strategies.

Again nobody talked about "preemptive changes in the trading direction as function of changes in fundamentals".

Pulling data makes sense when starting up strategies but generally the industry standard is a reactive architecture for handling data in a live trading environment and not pulling data.



Butterfly said:
A) same, a fundamental strategy has generally a long term view and therefore wouldn't react preemptively simply because one fundamental has changed. Fundamental data changes are not prices. Again, you are focusing on the wrong point.

B) I thought we were talking trading apps ? I didn't realize an analyst was a trading app. Incoming news processed by humans is not the issue here, it's how it's being used by the trading app. A fundamental strategy would not react on news or events simply because they are being pushed. Hence, a pull is more appropriate since the strategy will be more focus on a periodic cycle rather that an "ad hoc" cycle.
More...
 
I think you should actually go and read the post on the blog of the OP. then re-read what I wrote in my suggestions to OP. Nothing of what you talk about now is relevant to what OP elaborated on.

But hey, you hold your view that's fine. I respect it just totally disagree. Let's have both opinion stand on their own. I do not want to spend more time going back and forth with you because from past memory that does not go anywhere.

Butterfly said:
let's say your trading strategy is based purely on Earnings surprise or maybe actual Quarterly earnings change, both a fundamental data, the announcement will usually happen after the close. If the info was pushed through a "pipe" and your trading app capturing it, it couldn't act on it because the market would already be closed. A simple pull at the end of the day for "earnings" change could update your metrics table for the next morning trade.

let's say you missed the earnings news, would it be a tragedy ? if you are really long term fundamental, your trading strategy would rely on a number of factors, therefore a single data item change would not be enough to trigger a trade or change dramatically your portfolio structure. Further more, even if you caught the earnings news on time, it's more than likely than the market would have reacted already, hence making your trading decision pointless. It's also more likely than the market reaction will reverse soon in a few days after the over-reaction to the news is being "digested". I see this happens times and times again, like clockwork.

Either way, it shouldn't impact your trading strategy long term view of the securities you are holding for that strategy since other factors and a periodic cycle would only come into play for your real decision making process.
More...
 
the industry standard ? by who ? you ?

the industry standard has always been pull simply because push technology is actually recent (MQ, Events in OOP) in relative terms.

Again, speaking from experience, Portfolio Managers mostly use "pull" technology to get the data they need for their decision making process. They use FACTSET for example to pull reports and "views" of their portfolio holdings. Even in FACTSET Excel, it's pull technology, the OLE/DLL thing will do a regular query to FACTSET database server and populate the rows and columns of your spreadsheet. Nothing is pushed, and why should it be ?

I am assuming FACTSET is pretty much "standard" in the industry if you get my drift.
 
Dude we are talking algorithmic trading strategies. Where exactly did you start to get entirely lost?!?

Butterfly said:
the industry standard ? by who ? you ?

the industry standard has always been pull simply because push technology is actually recent (MQ, Events in OOP) in relative terms.

Again, speaking from experience, Portfolio Managers mostly use "pull" technology to get the data they need for their decision making process. They use FACTSET for example to pull reports and "views" of their portfolio holdings. Even in FACTSET Excel, it's pull technology, the OLE/DLL thing will do a regular query to FACTSET database server and populate the rows and columns of your spreadsheet. Nothing is pushed, and why should it be ?

I am assuming FACTSET is pretty much "standard" in the industry if you get my drift.
More...
 
volpunter said:
I think you should actually go and read the post on the blog of the OP. then re-read what I wrote in my suggestions to OP. Nothing of what you talk about now is relevant to what OP elaborated on.

But hey, you hold your view that's fine. I respect it just totally disagree. Let's have both opinion stand on their own. I do not want to spend more time going back and forth with you because from past memory that does not go anywhere.
More...

I think your experience is limited to sell-side firms, brokers or market-makers and HFT traders, where price is at the center of everything. It's really the top of the iceberg, the whole industry is far more diverse and voluminous than sell-side firms selling their trading venues by any means necessary.

Sell-side create liquidity, but they don't create wealth per se. Only a long term view of holding assets can create wealth, not a short term view. They are basically at the end of the food chain even though they tend to believe that they are at the top. They create market noise, and that's their only purpose, yet a very useful purpose I might add.
 
indeed, they are also systematic trading strategies

and they need pull data, not push

I am perfectly on topic despite your attempt to derail with your Analyst and news analogy, which I also answered
 
You must log in or register to reply here.
Back
Top