Commit Graph

28 Commits

Author SHA1 Message Date
Gerardo Salazar e2964dd4b1 Make OrderListHash deterministic by using MD5 as its underlying hash function (#5276)
* Update OrderListHash to use MD5 as hash instead of hash code

* Update regression algorithm OrderListHash statistic

* Use full MD5 hash as OrderListHash, update regression statistic

* Fixes failing regression tests
2021-02-09 12:19:25 -03:00
Juan José D'Ambrosio 765f2cab3b Add primary exchage to map files (#5194)
* Add main exchange to map file

minimum minimorum test

* Add MainExchange to MapFilRow IEquatable implementation
Also write CSV line correctly

* Normalize exchanges

Also add small test form MapFileRow

* Exchange name normalization

- Exchange name normalization.
- Tick will protobuf `ExchangeCode` which will set existing `Exchange`

* Include single character exchange code

* Encode exchange as byte, Include exchange name as Enum

* Update MapFileRow.cs to accept a byte instead a string for primary exchange

* Fix issues in encoding exchanges map

* Fix broken MapFile tests

* Rename Main Exchange property in MapFileRow

* Use BATS instead of BATSZ

* Add primary exchange from map file

* Move MapFilePrimaryExchangeProvider to its own file

* Implement Char for primary exchange in map files

* Fix broken tests after implementing char exchange into map file

* Add a new constructor for MapFileRow

it  accepts a PrimaryExxchange object as parameter

* Fix broken test 

WIP

* Address review 

WIP

* Handles unexpected encoded exchanges

* Keep addressing review

* Address review

* Fix broken tests

Make PrimaryExchange as default even if the IPrimaryExchangeProvider is not instantiated

* Revert breaking change.

* Add a check that resolved map file actually have content

The issue that leads to this change was caused by a map file resolved to APC in the CoarseFundamentalTop3Algorithm. 
The resolver returns a not-null MapFile, but it is empty.
To my best understand, the MapFile Resolver should return only existing map files in the map_files folder. But the LocalMapFileProvider returned a map file with a permtick that doesn't exist in the file system as CSV.

* Set PrimaryExchange.UNKNOWN as default

* Address review couple minor bugs

- Use SID market instead of USA
- Default ticke exchange is UNKNOWN

* Update map files with latest format and latest info

This include:
 - VXX, it was delisted, then VXXB appears and later it chnages from VXXB -> VXX
 - AAA delisting and a new listing with ticker AAA
 - FOXA ticker changue to TFCFA and later it was delisted

* Update test after updating latest map files

* Fix bug with factor files after updating map files

* Fix bug with factor files after updating map files

and revert statistics in regression tests

* Update OrderHash in regression test

Co-authored-by: Martin Molinero <martin.molinero1@gmail.com>
2021-01-21 18:46:12 -03:00
Martin-Molinero 91e8393aac DividedEventProvider distribution computation (#4828)
* DividedEventProvider distribution computation

- Update regression algorithm which was using a different reference
  price when calculating the dividend
- Adjust divided event provider to compute distribution using factor
  file reference price, if not 0. Adding unit tests
- For equities, only emit auxiliary data points for
  TradeBar configurations, not for QuoteBars, nor internal.

* Address reviews

- Split and Dividend event provider will throw an exception when there
  is no reference price available. Updating `wm` factor file which was
  missing references price and regression algorithms using WM.
- Updating unit tests asserting new exception
2020-11-11 15:47:51 -03:00
Martin Molinero 6ad123ad8c Update regression algorithms stats
- Update regression algorithms stats after making SecurityCache ignore
  QuoteBars for equity for OHCL values and GetLastData(). They were
  affected since the `BenchmarkSecurity` used `.Price` which was QB for
  equities. Order list hashes changed because SubmissionLastPrice will
  now be TB instead of QB
2020-04-08 19:31:42 -03:00
Martin Molinero 499248fe12 Revert "Revert "Adding Support for Equity L1 Quote Data""
This reverts commit 8cd8d206ca.
2020-04-06 10:32:59 -03:00
Martin Molinero 2a0b73e1cf Update regression algorithms
- Update regression algorithms after removing order based insights
2020-04-01 20:37:20 -03:00
Martin Molinero 486b2e6641 Update regression stats
- Update expected OrderListHash in regression algorithms due to
serialization improvements
2020-03-31 12:21:13 -03:00
Jack Simonson c6767780c2 Update regression test statistics 2020-03-26 18:29:29 -07:00
Jared 8cd8d206ca Revert "Adding Support for Equity L1 Quote Data" 2020-03-11 19:16:05 -07:00
Juan José D'Ambrosio 951e89314a Update Regression tests 2020-03-11 14:34:16 -03:00
Martin Molinero d5d00f8472 Update tests
- Update regression tests
- Add missing data
- Revert Messaging.cs change
- Update unit tests
2020-03-11 14:34:16 -03:00
D 3b25ea4240 Update test after adding the new sample data 2020-03-11 14:34:16 -03:00
Martin Molinero 93418d9e72 Fix rebase. Add test
- Fix rebase
- Update tests
- Add new unit test, that asserts fill model uses quote price is present
2020-03-11 14:34:13 -03:00
Juan José D'Ambrosio b7a503a096 Update Regression algorithms 2020-03-11 14:34:13 -03:00
Juan José D'Ambrosio fd9de5895a Consolidate only trades, low resolution reads trades only, update tests
Crypto and Equities will consolidate trades by default

Daily and hourly resolution will return only trades for crypto and equities. 
Update equity TAQ regression test 

Add minute sample files

Updating regression algorithms

Update SpotMarket test cases

Add regression test for equity trades and quotes
- History request.
- Trades and quotes pumped into OnData.
- Subscriptions are added correctly.

Add sample data

Checks low resolution only subscribes to trade bars
2020-03-11 14:34:05 -03:00
Martin Molinero 0027ab1e66 Regression test assert order list hash
- Regression tests assert order list hash value
- Normalizing some regression test behavior that had differences between
C# and Py
2020-02-19 20:26:12 -03:00
Martin Molinero 20c589f0fe Validate option universe selection data
- Remove unused `OptionChainUniverseDataCollectionAggregatorEnumerator`
- Refactor `BaseDataCollectionAggregatorEnumerator` and
`OptionChainUniverseDataCollectionEnumerator` to avoid emitting invalid
data points
- Updating regression tests statistics
- Adding unit test
2020-01-27 19:11:17 -03:00
Gerardo Salazar 61aa0d3a65 Updates regression statistics
Please note that with these changes, any algorithms that
use daily data exclusively will have incorrect statistics.
2020-01-16 14:03:54 -08:00
Martin Molinero cb4c574eeb Adjust algorithms statistics 2019-08-06 21:35:25 -03:00
Martin Molinero 40ac9c5033 Fix IB FeeModel Option fees
- Fixing IB FeeModel options fees. Adding unit tests
- Updating regression algorithms
2019-01-02 15:50:00 -03:00
Michael Handschuh 380caa5203 Add IRegressionAlgorithmDefinition.CanRunLocally
This flag indicates whether or not the local regression test system,
via RegressionTests.AlgorithmStatisticsRegression should run a given
IRegressionAlgorithmDefinition
2018-07-18 15:57:11 -04:00
Michael Handschuh 7eb4526724 Update open source data set and regression statistics
Factor files from 2018.06.04
Data   files from 2018.07.05
2018-07-05 15:13:57 -04:00
Martin Molinero 96dc5e8673 Moving IRegressionAlgorithmDefinition to QuantConnect.Common 2018-07-03 13:34:27 -03:00
Michael Handschuh 8402b6f01e Update factor files to 2018.06.04
It's important that we keep the factor files consistent with respect to
the date that they were generated. This enables us to run the regression
algorithms in the cloud and get the same results by using the factor files
from the correct date.
2018-06-07 12:16:45 -04:00
Michael Handschuh 9ee61f425c Refactor regression algorithm to IRegressionAlgorithmDefinition
A mechanical refactoring was performed to make algorithms currently used in
regression algorithms to implement IRegressionAlgorithmDefinition, which allows
algorithms to define their own expected statistics and what languages should be
run as part of regression. The type name of the  C# type is used to determine the
file/model name for python. This was for simplicity, but if needed, could later be
refactored to expose more information, but for now the convention of keeping names
the same makes sense and just works easily.
2018-06-05 12:10:50 -04:00
AlexCatarino 98a1b6a2a2 Refactors options algorithms
AddEquity and SetDataNormalization calls were removed because they are not mandatory since PostInitialize calls them.
2017-12-18 11:24:08 +00:00
Jared Broad fb7d1994ff Tidy and add tags 2017-09-20 17:14:44 -04:00
quant1729 5803115548 Option splits and renames in backtesting. Added regression tests. 2016-10-26 21:20:03 +02:00