{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "1c11b660", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[2023-10-25 10:45:37.856663] INFO: zipline.data.bundles.core: Ingesting tquant.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Merging daily equity files:\n" ] } ], "source": [ "import os \n", "StockList1 = ['1227', '1234', '1304', '1314', '1434', '1440', '1476', '1504', '1507', '1590', '1605', '1704', '1710', '1717', '1723', '1789',\n", " '1802', '1907', '2006', '2015', '2049', '2101', '2103', '2106', '2204', '2227', '2327', '2337', '2344', '2356', '2360', '2362',\n", " '2379', '2384', '2385', '2392', '2395', '2448', '2449', '2450', '2451', '2489', '2501', '2504', '2511', '2542', '2545', '2548',\n", " '2603', '2606', '2607', '2608', '2609', '2610', '2615', '2618', '2707', '2723', '2727', '2809', '2812', '2823', '2834', '2845',\n", " '2847', '2855', '2884', '2887', '2888', '2889', '2903', '2915', '3034', '3037', '3044', '3149', '3189', '3406', '3702', '4938',\n", " '4958', '5522', '5871', '6005', '6176', '6239', '6269', '6286', '8008', '8046', '8078', '8422', '9904', '9907', '9914', '9917', '9921',\n", " '9933', '9940', '9945', '2458', '5264', '2206', '1201', '2347', '3231', '5534', '6116', '9910', '1477', '2353', '6271', '1319',\n", " '1722', '2059', '3060', '3474', '3673', '2393', '2376', '2439', '3682', '1262', '2201', '2377', '3576', '2352', '2838', '8150',\n", " '2324', '2231', '8454', '2833', '6285', '6409', '1536', '1702', '2313', '2498', '2867', '6415', '6456', '9938', '2383', '4137', '6452',\n", " '1707', '1589', '2849', '6414', '8464', '2355', '2345', '3706', '2023', '2371', '1909', '2633', '3532', '9941', '2492', '3019',\n", " '3443', '4915', '4943', '1229', '2441', '2027', '3026', '1210', '2104', '2456', '5269', '8341', '2354', '3005', '3481', '6669',\n", " '2409', '3023', '6213', '2404', '3533', '6278', '6592', '3653', '3661', '3665', '2301', '3714', '2883', '2890', '6531', '1904',\n", " '2014', '2105', '2108', '2474', '2637', '6781', '1102', '4919', '1402', '3035', '3036', '4961', '6719', '6770', '2368', '1795',\n", " '6550', '6789', '3017', '1101', '1216', '1301', '1303', '1326', '2002', '2207', '2303', '2308', '2311', '2317', '2325', '2330',\n", " '2357', '2382', '2412', '2454', '2801', '2880', '2881', '2882', '2885', '2886', '2891', '2892', '2912', '3008', '3045', '3697',\n", " '4904', '5880', '6505', '2408', '3711', '5876']\n", "tickers = \" \".join(StockList1)\n", "\n", "os.environ['TEJAPI_BASE'] = 'https://api.tej.com.tw'\n", "os.environ['TEJAPI_KEY'] = 'Your key'\n", "\n", "os.environ['mdate'] = '20220101 20230702'\n", "os.environ['ticker'] = tickers\n", "\n", "!zipline ingest -b tquant\n", "\n", "from zipline.pipeline import Pipeline\n", "from zipline.TQresearch.tej_pipeline import run_pipeline\n", "from zipline.pipeline.data import TWEquityPricing\n", "from zipline.pipeline.factors import SimpleMovingAverage, AverageDollarVolume" ] }, { "cell_type": "markdown", "id": "6404f2d4", "metadata": {}, "source": [ "## Masking\n", "有時我們想在計算 pipeline 時忽略特定資產。在兩種常見情況下，忽略資產是很有用的：\n", "\n", "1. 我們想要計算一個計算成本很高的因子，並且我們只關心特定資產的計算結果。一個常見的範例是計算回歸係數（透過 **built-in factors** `RollingLinearRegressionOfReturns`）。\n", "\n", "\n", "2. 我們想要計算一個在不同資產之間比較的因子，但不希望針對所有資產進行比較。例如，我們可能希望使用 `Factor` 中的 `top` 方法來計算報酬率排名前 200 的資產，並忽略流動性太低的資產。\n", "\n", "為了實現這兩個情形，所有 `Factors` 和許多 `Factor` 方法都可以傳入 mask 參數，該參數必須是一個 `Filter` ，表示計算時要考慮哪些資產。" ] }, { "cell_type": "markdown", "id": "c085b120", "metadata": {}, "source": [ "### Masking Factors\n", "\n", "假設我們希望 pipeline 輸出具有高或低價格變動（`percent_difference`）的證券，但同時只考慮近30日成交金額超過 $ 1,000,000,000 元的證券。\n", "\n", "為此，我們重新建立 `make_pipeline` 函數，並建立 `high_dollar_volume` 過濾器。\n", "\n", "然後，我們可以透過將`high_dollar_volume`作為`mask`參數傳遞給`SimpleMovingAverage`，將此過濾器用作移動平均因子的`mask`。" ] }, { "cell_type": "code", "execution_count": 2, "id": "686d5a0a", "metadata": {}, "outputs": [], "source": [ "# Dollar volume factor\n", "dollar_volume = AverageDollarVolume(inputs = [TWEquityPricing.close, TWEquityPricing.volume],window_length=30)\n", "\n", "# High dollar volume filter\n", "high_dollar_volume = (dollar_volume > 1000000000)\n", "\n", "# Average close price factors\n", "mean_close_10 = SimpleMovingAverage(inputs=[TWEquityPricing.close], window_length=10, mask=high_dollar_volume)\n", "mean_close_30 = SimpleMovingAverage(inputs=[TWEquityPricing.close], window_length=30, mask=high_dollar_volume)\n", "\n", "# Relative difference factor\n", "percent_difference = (mean_close_10 - mean_close_30) / mean_close_30" ] }, { "cell_type": "markdown", "id": "b1432983", "metadata": {}, "source": [ "我們將 Mask 應用於 `SimpleMovingAverage`，限制平均收盤價因子的計算範圍為：滿足 `high_dollar_volume` filter 的 35 檔證券。而如果沒有 Mask 則會有 230 檔證券。\n", "\n", "當我們結合 `mean_close_10` 和 `mean_close_30` 來計算 `percent_difference` 時，該計算是在相同的 35 個證券上進行的。" ] }, { "cell_type": "markdown", "id": "35c1a4f4", "metadata": {}, "source": [ "### Masking Filters\n", "\n", "Masks 還可以使用不同 filters 來回傳結果，例如：`top`、`bottom`與`percentile_between`。\n", "\n", "Masks 在我們想要在建立投資組合時特別有用。例如：假設我們想要取得同時是收盤價前百分之十、開盤價前 8 名、成交值前百分之十的證券。\n", "\n", "我們可以透過以下方法達成。" ] }, { "cell_type": "code", "execution_count": 3, "id": "8a604ffb", "metadata": {}, "outputs": [], "source": [ "# Dollar volume factor\n", "dollar_volume = AverageDollarVolume(inputs = [TWEquityPricing.close, TWEquityPricing.volume],window_length=30)\n", "\n", "# High dollar volume filter\n", "high_dollar_volume = dollar_volume.percentile_between(90,100)\n", "\n", "# Top open price filter (high dollar volume securities)\n", "top_open_price = TWEquityPricing.open.latest.top(8, mask=high_dollar_volume)\n", "\n", "# Top percentile close price filter (high dollar volume, top 8 open price)\n", "high_close_price = TWEquityPricing.close.latest.percentile_between(90, 100, mask=top_open_price)" ] }, { "cell_type": "markdown", "id": "ac1b7418", "metadata": {}, "source": [ "讓我們將這些變數放入`make_pipeline`，接著輸出一個有著`high_close_price` filter 的空 pipeline \n", "\n" ] }, { "cell_type": "code", "execution_count": 4, "id": "713cd52d", "metadata": {}, "outputs": [], "source": [ "def make_pipeline():\n", "\n", " dollar_volume = AverageDollarVolume(inputs = [TWEquityPricing.close, TWEquityPricing.volume],window_length=30)\n", "\n", " high_dollar_volume = dollar_volume.percentile_between(90,100)\n", "\n", " top_open_price = TWEquityPricing.open.latest.top(8, mask=high_dollar_volume)\n", "\n", " high_close_price = TWEquityPricing.close.latest.percentile_between(90, 100, mask=top_open_price)\n", "\n", " return Pipeline(\n", " screen=high_close_price\n", " )" ] }, { "cell_type": "markdown", "id": "ce4bbf21", "metadata": {}, "source": [ "執行這個 pipeline " ] }, { "cell_type": "code", "execution_count": 5, "id": "fae5b3e2", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "通過過濾器的證券數: 1\n" ] }, { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
2022-12-30 00:00:00+00:00Equity(149 [3008])
\n", "
" ], "text/plain": [ "Empty DataFrame\n", "Columns: []\n", "Index: [(2022-12-30 00:00:00+00:00, Equity(149 [3008]))]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "result = run_pipeline(make_pipeline(), '2022-12-30', '2022-12-30')\n", "print('通過過濾器的證券數: %d' % len(result))\n", "result" ] }, { "cell_type": "markdown", "id": "f72a946d", "metadata": {}, "source": [ "像上述這樣應用 masks 的方式可以視為一個 \"asset funnel\"。" ] }, { "cell_type": "markdown", "id": "741aac93", "metadata": {}, "source": [ "### 比較Screen與Mask\n", "- screen：\n", " - 若使用`screen`，會在所有 pipeline factors 或 filters **計算完後**將不符合條件的資料刪除，以`high_close_price`這個 filter 為例，high_close_price=False 的公司將被刪除。\n", " - 若是遇到計算量很大的 factors 時這種做法會比較沒有效率。 \n", "\n", "- mask：\n", " - 若使用`mask`，會在該 factors 或 filters 計算時直接忽略 mask=False 的公司不納入計算。\n", " - 這種做法不需要等到輸出結果時才進行篩選，在 factors 或 filters 計算階段就已經進行篩選。" ] } ], "metadata": { "kernelspec": { "display_name": "Python [conda env:zipline-tej] *", "language": "python", "name": "conda-env-zipline-tej-py" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.13" } }, "nbformat": 4, "nbformat_minor": 5 }