-
Notifications
You must be signed in to change notification settings - Fork 1
Typical Use
To download prices during market hours:
bin/console price:sync -v --delay=random --chunk=50 --stillTsaveQ data/source/y_universe.csv
This command will handle the following scenarios as follows:
- Missing price history: download history from 2011-01-03 to T-1, then add current quote for T.
- Price history exists with last day earlier than T-1: download missing price history to T-1, then add current quote for T.
- Price history exists with last day at T-1: add current quote for T. For a case when you have price at T-1 as quote with timestamp earlier than market close for that day, use next command. This situation may occur if you were looking at charts mid-day and needed current quotes before the close of T. The command below will download price for T-1 as history, thus making sure you don't have non-closing price saved as history.
To download prices after market close as historical. You need to be aware when prices for a T become supplied by Yahoo API as historical. This time is after about 1800 EST (for NYC based exchanges such as NYSE and NASDAQ).
bin/console price:sync -v --delay=random --chunk=50 --prevT-QtoH data/source/y_universe.csv
This command will download historical prices if price at T-1 found to be a quote. Quotes in price history are represented with time component other than midnight.
This section will describe scanners that use OHLCV Entities.
Formulas work with scalar values and return scalar result.
Operands in formulas can be dealing with today's price data, or with data involving past values. For example:
- get closing price for today: "Close" or "Close(0)";
- get opening price 2 days ago: "Open(2)".
Operands in formulas can also represent so called aggregate values. These values are calculated as one scalar which involves several scalars. Example:
- 10 day moving average: "Average(10)";
- 20 day standard deviation: "Std_d(20)".
More complicated example would be to calculate a variance. Variance is determined by taking the difference between each given value and the average of all values given. Each of those differences is then squared, and the results are totaled. The average of that total is then determined. Variance formula can be written as follows: For 2 trading periods: sum((Close(0) - Average(2))^2 + (Close(1) - Average(2))^2)/2
In this system operands for getting elements of the OHLCV format are represented by custom functions in the Symfony's Expression Builder.
Simple formulas involve manipulation of fields readily available on entities, such as last Price, Open Price and so on.
Intermediate formulas handle entity fields X trading periods ago, and can be of form Close(3), Open(100), etc.
Advanced formulas handle mathematical functions such as Average, Sum, etc and can be more complex, thus involving several functions.
Price scans perform simple expression searches, conditions of which evaluate to boolean results. Examples:
- Find instruments where Last Closing Price > 100;
- Find instruments where Opening Price 3 days ago was below 200;
- Find instruments where 3 day Average of the Closing Price is above 25;
Price Formulas return values, or arrays of values.