getStats
The getStats
API is used for data aggregation and preprocessing. It is intended to run on a Data Access Process and aggregate records from a configured database table using a specified dictionary of arguments.
The gw
variable below is defined as an IPC connection to the Service Gateway. For example `:insights-qe-gateway:5050 would connect to the query environment gateway within an insights namespace.
Gateway URL
The gw
variable below is defined as an IPC connection to the Service Gateway. For example `:insights-qe-gateway:5050 would connect to the query environment gateway within an insights namespace.
// API arguments dictionary.
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`analytics; `firstPrice`lastPrice`avgPrice`sumPrice)
)
opts:()!()
last gw(`getStats;args;`callback;opts);
instrumentID firstPrice lastPrice avgPrice sumPrice
---------------------------------------------------
AMD 113.66 110.47 115.0438 68105.9
Gateway URL
The $INSIGHTS_URL
, $INSIGHTS_CLIENT_ID
, $INSIGHTS_CLIENT_SECRET
variables should point at your kdb Insights install.
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
analytics : ["firstPrice","lastPrice","avgPrice","sumPrice"]
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"instrumentID":"AMD","firstPrice":113.66,"lastPrice":110.47,"avgPrice":115.0438,"sumPrice":68105.9}]
When issuing an IPC request, synchronous requests return a tuple where the first element is the response payload and the second is the response data. See the asynchronous requests section for how to use the callback argument to get an asynchronous response. See the extended options section for how to use the options like timeouts for your IPC call.
Arguments
getStats
inherits all arguments from getTicks
except applyCanCor and columns, then has the following additional arguments.
name | required | type | default | example | description |
---|---|---|---|---|---|
analytics | no | symbol[] | () | `avgPrice | A list of symbols for analytics to apply. See the analytics section for more detail |
agg | no | list | () | ((`avg`price);(`first`price)) | Specifies the aggregations to perform. See agg section for supported formats |
groupBy | no | symbol[] | identifier column | `exch | List of columns to group aggregation result by |
granularity | no | symbol | 1 | 30 | Size of the time bucket for the aggregation. Used in conjunction with granularityUnit |
granularityUnit | no | symbol | N/A | `minute | Unit of the time bucket for the aggregation. Used in conjunction with granularity. Options are `second`minute`hour`day |
movingWindowSize | no | symbol | N/A | 60 | Optional modifier on granularity and granularityUnit. Makes the bucket size a lookback of this size backwards from each granularity bucket |
applyHolidayCalendar | no | boolean/symbol | 0b | `NYSE | Apply an exchange holiday calendar to racked/filled data to remove buckets on holidays |
analytics
There are number of named analytics available. These are built dynamically for all tables in the database. Generic operations (`first;`last) are applied to all columns, and numerical operations (`min;`max;`avg;`sum;`med) are applied where applicable. The naming convention is the aggregate keyword and the column to which it is applied. For example, the first, last, average and sum price;
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`analytics; `firstPrice`lastPrice`avgPrice`sumPrice)
)
last gw(`getStats;args;`callback;()!());
instrumentID firstPrice lastPrice avgPrice sumPrice
---------------------------------------------------
AMD 113.66 110.47 115.0438 68105.9
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
analytics : ["firstPrice","lastPrice","avgPrice","sumPrice"]
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"instrumentID":"AMD","firstPrice":113.66,"lastPrice":110.47,"avgPrice":115.0438,"sumPrice":68105.9}]
All the analytics that use `avg and `first for the Trade table are shown in the table below.
tableName | analytic | clause |
---|---|---|
Trade | firstEventTimestamp | *: `eventTimestamp |
Trade | firstInstrumentID | *: `instrumentID |
Trade | firstExchTime | *: `exchTime |
Trade | firstPrice | *: `price |
Trade | firstVolume | *: `volume |
Trade | firstConditions | *: `conditions |
Trade | firstNormalisedCondition | *: `normalisedCondition |
Trade | firstTickDirection | *: `tickDirection |
Trade | firstTradedExchange | *: `tradedExchange |
Trade | firstSequenceNumber | *: `sequenceNumber |
Trade | firstAccVol | *: `accVol |
Trade | firstTickCount | *: `tickCount |
Trade | firstExecutionID | *: `executionID |
Trade | firstSrcSys | *: `srcSys |
Trade | avgEventTimestamp | avg `eventTimestamp |
Trade | avgExchTime | avg `exchTime |
Trade | avgPrice | avg `price |
Trade | avgVolume | avg `volume |
Trade | avgSequenceNumber | avg `sequenceNumber |
Trade | avgAccVol | avg `accVol |
Trade | avgTickCount | avg `tickCount |
Complex and custom analytics can be added to getStats, see Customize getStats for more information.
agg
Free form aggregations or parse trees can be specified using the agg argument. The following formats are supported;
1. Dictionary in classic kdb column clause format
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; `avgPrice`firstPrice!((`avg;`price);(`first;`price)))
)
last gw(`getStats;args;`callback;()!())
instrumentID avgPrice firstPrice
--------------------------------
AMD 115.0438 113.66
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
agg : { "avgPrice":["avg","price"], "firstPrice":["first","price"] }
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"instrumentID":"AMD","avgPrice":115.0438,"firstPrice":113.66}]
Aggregations can be specified as symbols, strings or a combination of both. REST calls must be specified as strings.
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; `avgPrice`firstPrice!(("avg";`price);(`first;`price)))
)
last gw(`getStats;args;`callback;()!())
instrumentID avgPrice firstPrice
--------------------------------
AMD 115.0438 113.66
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; `avgPrice`somePriceCalc!((`avg;`price);(`abs;(`wavg;`price;(`abs;`volume)))))
)
last gw(`getStats;args;`callback;()!())
instrumentID avgPrice somePriceCalc
-----------------------------------
AMD 115.0438 49918.54
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
agg : { "avgPrice":["avg","price"], "somePriceCalc":["abs",["wavg","price",["abs","volume"]]] }
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"instrumentID":"AMD","avgPrice":115.0438,"somePriceCalc":49918.54}]
2. Single parse tree format that auto names columns as aggregation
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; ((`avg`price);(`first`price)))
)
last gw(`getStats;args;`callback;()!())
instrumentID aggregation1 aggregation2
--------------------------------------
AMD 115.0438 113.66
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
agg : [["avg","price"], ["first","price"]]
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"instrumentID":"AMD","aggregation1":115.0438,"aggregation2":113.66}]
3. Triplet symlist (outputColumnName;operator;column)
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; ((`c1`avg`price);(`c2`first`price)))
)
last gw(`getStats;args;`callback;()!())
instrumentID c1 c2
----------------------------
AMD 115.0438 113.66
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
agg : [["c1","avg","price"], ["c2","first","price"]]
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"instrumentID":"AMD","c1":115.0438,"c2":113.66}]
groupBy
Defaults to the identifier column, which is defined as the left-most sum column in the table schema (instrumentID
). For example, group by the traded exchange;
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; ((`c1`avg`price);(`c2`first`price)));
(`groupBy; `tradedExchange)
)
last gw(`getStats;args;`callback;()!())
tradedExchange c1 c2
------------------------------
115.0438 113.66
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
agg : [["c1","avg","price"], ["c2","first","price"]],
groupBy : "tradedExchange"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"tradedExchange":"","c1":115.0438,"c2":113.66}]
granularity/granularityUnit
These arguments allow the user to aggregate over specified time buckets, the default is to aggregate across the entire data set. The start time of the first bucket is startTS
. For example, hourly buckets;
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`agg; `avgPrice`firstPrice!((`avg;`price);(`first;`price)));
(`granularity; 1);
(`granularityUnit; `hour)
)
last gw(`getStats;args;`callback;()!())
eventTimestamp instrumentID avgPrice firstPrice
--------------------------------------------------------------
2023.07.21D00:00:00.000000000 AMD 115.3425 113.66
2023.07.21D01:00:00.000000000 AMD 116.6158 118.89
2023.07.21D02:00:00.000000000 AMD 115.7733 114.94
2023.07.21D03:00:00.000000000 AMD 114.365 118.3
2023.07.21D04:00:00.000000000 AMD 114.915 115.22
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
agg : { "avgPrice":["avg","price"], "firstPrice":["first","price"] },
granularity : "1",
granularityUnit : "hour"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"eventTimestamp":"2023-07-21T00:00:00.000000000","instrumentID":"AMD","avgPrice":115.3425,"firstPrice":113.66},{"eventTimestamp":"2023-07-21T01:00:00.000000000","instrumentID":"AMD","avgPrice":116.6158,"firstPrice":118.89},...
movingWindowSize
The movingWindowSize
argument is a modifier on granularity/granularityUnit
, that makes the bucket size a lookback of the value specfied backwards from each bucket. Example trade data is used below to illustrate this feature;
args: (!) . flip (
(`table; `trade);
(`startTS; 2023.07.27D00:00:00.000000000);
(`endTS; 2023.07.28D00:00:00.000000000)
);
last gw(`getTicks;args;`callback;()!())
timecolumn idcolumn price
--------------------------------------------
2023.07.27D00:00:00.000000000 A 0
2023.07.27D00:48:00.000000000 A 1
2023.07.27D01:36:00.000000000 A 2
2023.07.27D02:24:00.000000000 A 3
2023.07.27D03:12:00.000000000 A 4
2023.07.27D04:00:00.000000000 A 5
2023.07.27D04:48:00.000000000 A 6
2023.07.27D05:36:00.000000000 A 7
2023.07.27D06:24:00.000000000 A 8
2023.07.27D07:12:00.000000000 A 9
2023.07.27D08:00:00.000000000 A 10
2023.07.27D08:48:00.000000000 A 11
2023.07.27D09:36:00.000000000 A 12
2023.07.27D10:24:00.000000000 A 13
2023.07.27D11:12:00.000000000 A 14
2023.07.27D12:00:00.000000000 A 15
2023.07.27D12:48:00.000000000 A 16
2023.07.27D13:36:00.000000000 A 17
2023.07.27D14:24:00.000000000 A 18
2023.07.27D15:12:00.000000000 A 19
2023.07.27D16:00:00.000000000 A 20
2023.07.27D16:48:00.000000000 A 21
2023.07.27D17:36:00.000000000 A 22
2023.07.27D18:24:00.000000000 A 23
2023.07.27D19:12:00.000000000 A 24
2023.07.27D20:00:00.000000000 A 25
2023.07.27D20:48:00.000000000 A 26
2023.07.27D21:36:00.000000000 A 27
2023.07.27D22:24:00.000000000 A 28
2023.07.27D23:12:00.000000000 A 29
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getTicks" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "trade",
startTS : "2023.07.27D00:00:00",
endTS : "2023.07.28D00:00:00"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"timecolumn":"2023-07-27T00:00:00.000000000","idcolumn":"A","price":0},{"timecolumn":"2023-07-27T00:48:00.000000000","idcolumn":"A","price":1},...
A getStats
call for the day with 4 hour buckets, shows that the bucket timestamps are the start of the buckets.
args: (!) . flip (
(`table; `trade);
(`startTS; 2023.07.23D00:00:00.000000000);
(`endTS; 2023.07.24D00:00:00.000000000);
(`analytics; `firstPrice`lastPrice`avgPrice);
(`granularity; 4);
(`granularityUnit; `hour)
);
last gw(`getStats;args;`callback;()!());
idcolumn timecolumn firstPrice lastPrice avgPrice
--------------------------------------------------------------------
A 2023.07.27D00:00:00.000000000 0 4 2
A 2023.07.27D04:00:00.000000000 5 9 7
A 2023.07.27D08:00:00.000000000 10 14 12
A 2023.07.27D12:00:00.000000000 15 19 17
A 2023.07.27D16:00:00.000000000 20 24 22
A 2023.07.27D20:00:00.000000000 25 29 27
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "trade",
startTS : "2023.07.27D00:00:00",
endTS : "2023.07.28D00:00:00",
analytics : ["firstPrice","lastPrice","avgPrice"],
granularity : "4",
granularityUnit : "hour"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"idcolumn":"A","timecolumn":"2023-07-27T00:00:00.000000000","firstPrice":0,"lastPrice":4,"avgPrice":2},{"idcolumn":"A","timecolumn":"2023-07-27T04:00:00.000000000","firstPrice":5,"lastPrice":9,"avgPrice":7},...
A movingWindowSize
that is the same as the granularity results in the same behavior as if movingWindowSize
was not provided, with the exception that the timestamps are the bucket ends.
args: (!) . flip (
(`table; `trade);
(`startTS; 2023.07.27D00:00:00.000000000);
(`endTS; 2023.07.28D00:00:00.000000000);
(`analytics; `firstPrice`lastPrice`avgPrice);
(`granularity; 4);
(`granularityUnit; `hour);
(`movingWindowSize; 4)
);
last gw(`getStats;args;`callback;()!());
timecolumn idcolumn firstPrice lastPrice avgPrice
--------------------------------------------------------------------
2023.07.27D04:00:00.000000000 A 0 4 2
2023.07.27D08:00:00.000000000 A 5 9 7
2023.07.27D12:00:00.000000000 A 10 14 12
2023.07.27D16:00:00.000000000 A 15 19 17
2023.07.27D20:00:00.000000000 A 20 24 22
2023.07.28D00:00:00.000000000 A 25 29 27
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "trade",
startTS : "2023.07.27D00:00:00",
endTS : "2023.07.28D00:00:00",
analytics : ["firstPrice","lastPrice","avgPrice"],
granularity : "4",
granularityUnit : "hour",
movingWindowSize: "4"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"timecolumn":"2023-07-27T04:00:00.000000000","idcolumn":"A","firstPrice":0,"lastPrice":4,"avgPrice":2},{"timecolumn":"2023-07-27T08:00:00.000000000","idcolumn":"A","firstPrice":5,"lastPrice":9,"avgPrice":7},...
A movingWindowSize
that is the not the same as the granularity returns the same buckets, but the aggregations are now performed over the lookback period. For example, the below getStats call performs a lookback 8 hours every 4 hours.
args: (!) . flip (
(`table; `trade);
(`startTS; 2023.07.27D00:00:00.000000000);
(`endTS; 2023.07.28D00:00:00.000000000);
(`analytics; `firstPrice`lastPrice`avgPrice);
(`granularity; 4);
(`granularityUnit; `hour);
(`movingWindowSize; 8)
);
last gw(`getStats;args;`callback;()!());
timecolumn idcolumn firstPrice lastPrice avgPrice
--------------------------------------------------------------------
2023.07.27D04:00:00.000000000 A 0 4 2
2023.07.27D08:00:00.000000000 A 0 9 4.5
2023.07.27D12:00:00.000000000 A 5 14 9.5
2023.07.27D16:00:00.000000000 A 10 19 14.5
2023.07.27D20:00:00.000000000 A 15 24 19.5
2023.07.28D00:00:00.000000000 A 20 29 24.5
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "trade",
startTS : "2023.07.27D00:00:00",
endTS : "2023.07.28D00:00:00",
analytics : ["firstPrice","lastPrice","avgPrice"],
granularity : "4",
granularityUnit : "hour",
movingWindowSize: "8"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"timecolumn":"2023-07-27T04:00:00.000000000","idcolumn":"A","firstPrice":0,"lastPrice":4,"avgPrice":2},{"timecolumn":"2023-07-27T08:00:00.000000000","idcolumn":"A","firstPrice":0,"lastPrice":9,"avgPrice":4.5},...
Your bucket end times are always the same regardless of what your movingWindowSize
is. For example, asking to lookback 3 hours every 4 hours, the bucket times would still be 4 hourly.
args: (!) . flip (
(`table; `trade);
(`startTS; 2023.07.27D00:00:00.000000000);
(`endTS; 2023.07.28D00:00:00.000000000);
(`analytics; `firstPrice`lastPrice`avgPrice);
(`granularity; 4);
(`granularityUnit; `hour);
(`movingWindowSize; 3)
);
last gw(`getStats;args;`callback;()!());
timecolumn idcolumn firstPrice lastPrice avgPrice
--------------------------------------------------------------------
2023.07.27D04:00:00.000000000 A 2 4 3
2023.07.27D08:00:00.000000000 A 7 9 8
2023.07.27D12:00:00.000000000 A 12 14 13
2023.07.27D16:00:00.000000000 A 17 19 18
2023.07.27D20:00:00.000000000 A 22 24 23
2023.07.28D00:00:00.000000000 A 27 29 28
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "trade",
startTS : "2023.07.27D00:00:00",
endTS : "2023.07.28D00:00:00",
analytics : ["firstPrice","lastPrice","avgPrice"],
granularity : "4",
granularityUnit : "hour",
movingWindowSize: "3"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"timecolumn":"2023-07-27T04:00:00.000000000","idcolumn":"A","firstPrice":2,"lastPrice":4,"avgPrice":3},{"timecolumn":"2023-07-27T08:00:00.000000000","idcolumn":"A","firstPrice":7,"lastPrice":9,"avgPrice":8},...
A common use case is a movingWindowSize of 7, a granularity of 1, and a granularityUnit of day that would produce a 7 day moving average each day.
Note
Only data within the bounds of startTS/endTS is included in the aggregation.
applyHolidayCalendar
Argument to be used in conjunction with fill and granularityUnit. Its purpose is to remove daily buckets of holidays when racking and filling, such that linear interpolation ignores weekends/holidays. Requires ExchangeHolidayCal
reference data to work.
Options
- 0b - does nothing.
- 1b - Will lookup and match tradedExchange in the target data to the exchangeID in the holiday calendar, and apply on a per exchange basis.
- Name of exchange as a symbol. This applies the holidays of exchangeID in the holiday calendar to all data, regardless of listed tradedExchange. This can be useful if the target data has not got an exchange listed. Also this usage enables a null symbol ` to be provided, if there is no exchangeID in the holiday calendar, this applies everything in the calendar.
args: (!) . flip (
(`table; `Trade);
(`startTS; 2023.07.21D00:00:00);
(`endTS; 2023.07.22D00:00:00);
(`analytics; `avgPrice`maxPrice);
(`granularity; 1);
(`granularityUnit; `hour);
(`fill; `linear);
(`applyHolidayCalendar; 1b)
)
last gw(`getStats;args;`callback;()!())
eventTimestamp instrumentID maxPrice avgPrice
------------------------------------------------------------
2023.07.21D00:00:00.000000000 AMD 118.99 115.3425
2023.07.21D01:00:00.000000000 AMD 119.6 116.6158
2023.07.21D02:00:00.000000000 AMD 119.19 115.7733
2023.07.21D03:00:00.000000000 AMD 119.16 114.365
2023.07.21D04:00:00.000000000 AMD 118.85 114.915
..
INSIGHTS_TOKEN=`curl -s --header "Content-Type: application/x-www-form-urlencoded" \
-d "grant_type=client_credentials&client_id=$INSIGHTS_CLIENT_ID&client_secret=$INSIGHTS_CLIENT_SECRET" \
"${INSIGHTS_URL}/auth/realms/insights/protocol/openid-connect/token" | jq -cr ".access_token"`
DATA=`curl -s -X POST "${INSIGHTS_URL}/servicegateway/fsi/getStats" \
-H "Content-Type: application/json" \
-H "Accept: application/json" \
-H "Authorization: Bearer $INSIGHTS_TOKEN" \
-d "$(jq -n \
'{
table : "Trade",
startTS : "2023.07.21D00:00:00",
endTS : "2023.07.22D00:00:00",
analytics : ["avgPrice","maxPrice"],
granularity : "1",
granularityUnit : "hour",
fill : "linear",
applyHolidayCalendar : "1b"
}' | jq -cr .)"`
echo $DATA | jq -cr '.payload'
[{"eventTimestamp":"2023-07-21T00:00:00.000000000","instrumentID":"AMD","maxPrice":118.99,"avgPrice":115.3425},{"eventTimestamp":"2023-07-21T01:00:00.000000000","instrumentID":"AMD","maxPrice":119.6,"avgPrice":116.6158},...