{ "cells": [ { "cell_type": "code", "execution_count": 21, "id": "e706cfb0-2234-4c4c-95d8-d1968f656aa0", "metadata": { "tags": [] }, "outputs": [], "source": [ "from datetime import datetime\n", "import pandas as pd\n", "from tms_data_interface import SQLQueryInterface\n", "\n", "query = \"\"\"\n", "WITH \n", "-- Capture all orders and trades within the spoofing time window\n", "trade_window AS (\n", " SELECT\n", " t.trade_id,\n", " t.trader_id,\n", " t.date_time AS trade_time,\n", " t.trade_side,\n", " t.trade_volume,\n", " o.trader_id AS order_trader_id,\n", " o.date_time AS order_time,\n", " o.order_volume,\n", " o.order_status,\n", " o.order_price,\n", " o.side AS order_side\n", " FROM \n", " {trade_data_1b} t\n", " LEFT JOIN \n", " order_10m o ON o.date_time BETWEEN t.date_time - INTERVAL '{spoofing_time_window_s}' SECOND \n", " AND t.date_time\n", " WHERE \n", " o.side = '{spoofing_side}'\n", "),\n", "\n", "-- Calculate net order volume for the specific trader\n", "net_order_volume_cte AS (\n", " SELECT \n", " trader_id,\n", " trade_id,\n", " trade_time,\n", " SUM(CASE \n", " WHEN order_status = 'new' THEN order_volume \n", " WHEN order_status = 'cancelled' THEN -order_volume \n", " WHEN order_status = 'fulfilled' THEN -order_volume \n", " ELSE 0 \n", " END) AS net_order_volume,\n", " COUNT(*) AS num_orders\n", " FROM trade_window\n", " WHERE order_trader_id = trader_id -- Filter by the trader who executed the trade\n", " GROUP BY trader_id, trade_id, trade_time\n", "),\n", "\n", "-- Calculate total net order volume for all traders (i.e., for spoofing side orders)\n", "net_order_volume_all_cte AS (\n", " SELECT \n", " trade_id,\n", " SUM(CASE \n", " WHEN order_status = 'new' THEN order_volume \n", " WHEN order_status = 'cancelled' THEN -order_volume \n", " WHEN order_status = 'fulfilled' THEN -order_volume \n", " ELSE 0 \n", " END) AS net_order_volume_all\n", " FROM trade_window\n", " GROUP BY trade_id\n", "),\n", "\n", "-- Calculate total trade volume on the opposite side (e.g., sell if spoofing is on buy)\n", "opposite_trade_volume_cte AS (\n", " SELECT \n", " t.trader_id,\n", " t.trade_id,\n", " SUM(t.trade_volume) AS total_trade_volume\n", " FROM {trade_data_1b} t\n", " WHERE \n", " t.date_time BETWEEN t.date_time - INTERVAL '{trade_time_window_s}' SECOND\n", " AND t.date_time\n", " AND t.trade_side = CASE WHEN '{spoofing_side}' = 'buy' THEN 'sell' ELSE 'buy' END\n", " GROUP BY t.trader_id, t.trade_id\n", ")\n", "\n", "-- Final result with calculated spoofing indicators\n", "SELECT\n", " n.trade_id,\n", " n.trader_id,\n", " n.trade_time,\n", " n.num_orders,\n", " n.net_order_volume,\n", " CASE \n", " WHEN o.total_trade_volume > 0 THEN n.net_order_volume / o.total_trade_volume\n", " ELSE NULL\n", " END AS order_trade_ratio,\n", " CASE \n", " WHEN a.net_order_volume_all > 0 THEN n.net_order_volume / a.net_order_volume_all\n", " ELSE NULL\n", " END AS volume_percentage\n", "FROM \n", " net_order_volume_cte n\n", "LEFT JOIN \n", " opposite_trade_volume_cte o ON n.trade_id = o.trade_id\n", "LEFT JOIN \n", " net_order_volume_all_cte a ON n.trade_id = a.trade_id\n", "WHERE \n", " n.net_order_volume > 0 -- Only consider positive net order volumes (potential spoofing);\n", " limit 1000\n", "\"\"\"\n", "\n", "class Scenario:\n", " seq = SQLQueryInterface(schema=\"internal\")\n", "\n", " def logic(self, **params):\n", " spoofing_time_window = params.get('spoofing_time_window', 300000) # default to 300,000 ms (5 minutes)\n", " spoofing_side = params.get('spoofing_side', 'buy')\n", " use_volume_for_order_trade_ratio = params.get('use_volume_for_order_trade_ratio', True)\n", " trade_time_window = params.get('trade_time_window', 300000)\n", " ignore_trade_after_spoofing = params.get('ignore_trade_after_spoofing', True)\n", " ignore_price_improvement = params.get('ignore_price_improvement', True)\n", "\n", " # Convert time windows from milliseconds to seconds\n", " spoofing_time_window_s = int(spoofing_time_window / 1000)\n", " trade_time_window_s = int(trade_time_window / 1000)\n", "\n", " query_start_time = datetime.now()\n", " print(\"Query start time:\", query_start_time)\n", "\n", " # Execute the query with the parameters passed from `params`\n", " row_list = self.seq.execute_raw(query.format(\n", " trade_data_1b=\"trade_10m_v3\", # Replace with actual table name\n", " spoofing_time_window_s=spoofing_time_window_s,\n", " trade_time_window_s=trade_time_window_s,\n", " spoofing_side=spoofing_side\n", " ))\n", "\n", " # Define columns for the resulting DataFrame\n", " cols = [\n", " 'trade_id', 'focal_id', 'trade_time', 'num_orders', \n", " 'net_order_volume', 'order_trade_ratio', 'volume_percentage'\n", " ]\n", "\n", " # Create a DataFrame from the query result\n", " final_scenario_df = pd.DataFrame(row_list, columns=cols)\n", "\n", "\n", " # Adding additional columns\n", " final_scenario_df['segment'] = 'Default'\n", " final_scenario_df['sar_flag'] = 'N'\n", " final_scenario_df['risk'] = 'Low Risk'\n", "\n", " return final_scenario_df" ] }, { "cell_type": "code", "execution_count": 22, "id": "b5c4307f-bc35-47e2-b680-fd1df2168d6c", "metadata": { "tags": [] }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Query start time : 2024-10-14 07:40:43.846637\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
focal_IDtrade_time_windownet_volumeorder_counttotal_trade_volumeorder_trade_ratiovolume_percentageSegmentSAR_FLAGRisk
030977282072024-01-01 00:03:00-92.0192-1.00.0DefaultNLow Risk
132286453222024-01-01 00:06:00-689.01689-1.00.0DefaultNLow Risk
227018727272024-01-01 00:09:00-42.0142-1.00.0DefaultNLow Risk
316590566552024-01-01 00:11:00-167.01167-1.00.0DefaultNLow Risk
416612888872024-01-01 00:13:00-756.01756-1.00.0DefaultNLow Risk
.................................
9519457726822024-01-01 00:43:00-854.01854-1.00.0DefaultNLow Risk
9621374780412024-01-01 00:43:00-926.01926-1.00.0DefaultNLow Risk
9771383291642024-01-01 00:43:00-433.01433-1.00.0DefaultNLow Risk
9818670074412024-01-01 00:43:00-626.01626-1.00.0DefaultNLow Risk
9923479063492024-01-01 00:43:00-69.0169-1.00.0DefaultNLow Risk
\n", "

100 rows × 10 columns

\n", "
" ], "text/plain": [ " focal_ID trade_time_window net_volume order_count \\\n", "0 3097728207 2024-01-01 00:03:00 -92.0 1 \n", "1 3228645322 2024-01-01 00:06:00 -689.0 1 \n", "2 2701872727 2024-01-01 00:09:00 -42.0 1 \n", "3 1659056655 2024-01-01 00:11:00 -167.0 1 \n", "4 1661288887 2024-01-01 00:13:00 -756.0 1 \n", ".. ... ... ... ... \n", "95 1945772682 2024-01-01 00:43:00 -854.0 1 \n", "96 2137478041 2024-01-01 00:43:00 -926.0 1 \n", "97 7138329164 2024-01-01 00:43:00 -433.0 1 \n", "98 1867007441 2024-01-01 00:43:00 -626.0 1 \n", "99 2347906349 2024-01-01 00:43:00 -69.0 1 \n", "\n", " total_trade_volume order_trade_ratio volume_percentage Segment \\\n", "0 92 -1.0 0.0 Default \n", "1 689 -1.0 0.0 Default \n", "2 42 -1.0 0.0 Default \n", "3 167 -1.0 0.0 Default \n", "4 756 -1.0 0.0 Default \n", ".. ... ... ... ... \n", "95 854 -1.0 0.0 Default \n", "96 926 -1.0 0.0 Default \n", "97 433 -1.0 0.0 Default \n", "98 626 -1.0 0.0 Default \n", "99 69 -1.0 0.0 Default \n", "\n", " SAR_FLAG Risk \n", "0 N Low Risk \n", "1 N Low Risk \n", "2 N Low Risk \n", "3 N Low Risk \n", "4 N Low Risk \n", ".. ... ... \n", "95 N Low Risk \n", "96 N Low Risk \n", "97 N Low Risk \n", "98 N Low Risk \n", "99 N Low Risk \n", "\n", "[100 rows x 10 columns]" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" } ], "source": [ "scenario = Scenario()\n", "scenario.logic(validation_window=300000)" ] }, { "cell_type": "code", "execution_count": null, "id": "36b1b24a-aeca-4d22-a2b3-6e04aca31695", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.8" } }, "nbformat": 4, "nbformat_minor": 5 }