{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":84493,"databundleVersionId":9871156,"sourceType":"competition"}],"dockerImageVersionId":30786,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# Introduction\n\nThis is a proof of concept of how to address this challenge using online learning. I wrote a blog post about it [https://serjhenrique.com/arima-and-online-learning-in-financial-forecasting/](https://serjhenrique.com/arima-and-online-learning-in-financial-forecasting/).\n\nFor those who don't want to read the full post the following extraction can give a good idea of the work done:\n\nFrom this [notebook](https://www.kaggle.com/code/chumajin/janestreet-easy-to-understand-new-time-series-api/notebook) we understood that new data is served by time_id. In other words, we receive a batch related from time_id = 0 and must return the predictions for this batch. According to the competition host, the lags parameter will be provided every time time_id = 0. Therefore, whenever the lags are not null, we can assume it is a new date_id. Based on this information, I chose to train the ARIMA model each time we encounter a new date_id.\n\nThe main idea is to train an ARIMA model for each symbol_id whenever we have a new date_id. We predict an entire day (968 time units) for every symbol and store these predictions in the Jane Predictor object. Then, for every prediction call that doesn't have lags (meaning we are still within the same already predicted date_id), we retrieve the first prediction for each symbol from our Jane Predictor's symbol_arr and pred_arr, remove it from the Jane Predictor's attributes, join it with the test DataFrame, and return it as the prediction. In other words, symbol_arr and pred_arr function like a stack that we consume with each new time_id until we reach a new date_id.\n\n![image.png](attachment:3de2b77b-cb20-4ac9-a8ec-b3ba56c88eb6.png)\n\nI hope you enjoy!!","metadata":{},"attachments":{"3de2b77b-cb20-4ac9-a8ec-b3ba56c88eb6.png":{"image/png":"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"},"76dea0a8-8a22-46a6-a8e7-a4324b2a636a.png":{"image/png":"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"}}},{"cell_type":"code","source":"import polars as pl\nimport pandas as pd\nimport numpy as np\nfrom typing import List, Tuple\nimport gc\nfrom abc import ABC, abstractmethod\n\nimport pyarrow.parquet as pq\n\nimport kaggle_evaluation.jane_street_inference_server\n\nimport os\nimport shutil\nimport glob\n\nimport multiprocessing\nfrom multiprocessing import Pool\nimport functools\n\nimport matplotlib.pyplot as plt\n\nfrom statsmodels.tsa.arima.model import ARIMA","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2024-12-02T20:24:26.069033Z","iopub.execute_input":"2024-12-02T20:24:26.069619Z","iopub.status.idle":"2024-12-02T20:24:26.076700Z","shell.execute_reply.started":"2024-12-02T20:24:26.069576Z","shell.execute_reply":"2024-12-02T20:24:26.075508Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class ArimaTrainer():\n    def __init__(\n            self,\n            train_size: int, # in days\n            data: pd.DataFrame = None,\n        ):\n        \"\"\"\n        Initialize the ArimaTrainer class with training size and data.\n\n        Args:\n            train_size (int): Number of days to be used for training.\n            data (pd.DataFrame, optional): DataFrame containing the data to be used. Defaults to None.\n        \"\"\"\n\n        self.train_size = train_size\n        self.data = data\n        self.unique_dates = []\n\n    def get_last_n_dates_per_symbol(self, data: pd.DataFrame, train_size: int) -> pd.DataFrame:\n        \"\"\"\n        Load data from a DataFrame and return the last n dates for each symbol_id\n        \n        Args:\n            data: Pandas DataFrame with training data\n            train_size: Number of distinct dates to keep per symbol\n            \n        Returns:\n            pd.DataFrame: Filtered DataFrame with last n dates per symbol\n        \"\"\"\n        \n        # Get unique dates per symbol\n        unique_dates = data[['symbol_id', 'date_id']].drop_duplicates()\n        \n        # Sort by symbol_id and date_id in descending order, then group and take top N dates\n        top_dates = (\n            unique_dates\n            .sort_values(by=['symbol_id', 'date_id'], ascending=[True, False])\n            .groupby('symbol_id')\n            .head(train_size)\n        )\n        \n        # Filter original data using these dates\n        final_result = (\n            data\n            .merge(top_dates, on=['symbol_id', 'date_id'], how='inner')\n            .sort_values(by=['symbol_id', 'date_id', 'time_id'])\n            [['date_id', 'time_id', 'symbol_id', 'responder_6', 'weight']]\n        )\n        \n        # Validate that each symbol has exactly train_size distinct dates\n        date_counts = final_result.groupby(\"symbol_id\")['date_id'].nunique().reset_index(name='distinct_date_count')\n        \n        assert (date_counts['distinct_date_count'] == train_size).all(), (\n            f\"Not all symbols have exactly {train_size} distinct dates. \"\n            f\"Found counts: {date_counts[date_counts['distinct_date_count'] != train_size]}\"\n        )\n        \n        return final_result\n\n    def _train_and_predict(self, s):\n        \"\"\"\n        Train an ARIMA model for a given symbol and make predictions.\n\n        Args:\n            s: Symbol identifier for which the model is trained.\n\n        Returns:\n            tuple: A tuple containing arrays of symbols and predictions.\n        \"\"\"\n        print(f'Starting training of the symbol {s}\\n')\n        \n        df_train = self.data.loc[\n            self.data['symbol_id'] == s\n        ]\n\n        if len(df_train) > 0:\n            df_train = self.get_last_n_dates_per_symbol(df_train, self.train_size).reset_index(drop=True)\n            model = ARIMA(df_train['responder_6'], order=(1, 0, 0))\n            model_fit = model.fit()\n\n            # After date_id 677 the time units per day stabilizes in 968\n            y_pred = model_fit.forecast(steps=968).astype(np.float32).to_numpy() \n            symbol = np.full(len(y_pred), s)\n        else:\n            y_pred = np.full(968, 0.0).astype(np.float32)\n            symbol = np.full(len(y_pred), s)\n        \n        return symbol, y_pred\n\n    def run(self):\n        \"\"\"\n        Execute the ARIMA training and prediction process for all symbols.\n\n        Returns:\n            tuple: Two concatenated arrays containing symbols and predictions.\n        \"\"\"\n        unique_symbols = self.data['symbol_id'].unique()\n        \n        # Use multiprocessing to parallelize the computation\n        with multiprocessing.Pool() as pool:\n            results = list(pool.map(self._train_and_predict, unique_symbols))\n\n        # Unpack the results into two separate lists\n        symbol, pred = zip(*results)\n        symbol = np.concatenate(symbol)\n        pred = np.concatenate(pred)\n\n        return symbol, pred\n            \n        \n            ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T20:24:26.079051Z","iopub.execute_input":"2024-12-02T20:24:26.079535Z","iopub.status.idle":"2024-12-02T20:24:26.094830Z","shell.execute_reply.started":"2024-12-02T20:24:26.079469Z","shell.execute_reply":"2024-12-02T20:24:26.093391Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"class JanePredictor():\n    def __init__(\n        self,\n        initial_data: pl.DataFrame,\n        train_size: int\n    ):\n        \"\"\"\n        Initializes the JanePredictor class with initial data and training size.\n\n        Parameters:\n        initial_data (pl.DataFrame): The initial dataset to be used for predictions.\n        train_size (int): The size of the training dataset.\n\n        Attributes:\n        lags_ (None): Placeholder for lag data.\n        cached_test_data (pl.DataFrame): Stores the initial data for caching purposes.\n        train_size (int): Stores the size of the training data.\n        \"\"\"\n        self.lags_ = None\n        self.cached_test_data = initial_data\n        self.train_size = train_size\n        \n    \n    def predict(self, test: pl.DataFrame, lags: pl.DataFrame | None) -> pl.DataFrame | pd.DataFrame:\n        \"\"\"\n        Predicts future values based on the test data and optional lag data.\n\n        Parameters:\n        test (pl.DataFrame): The test dataset containing new observations.\n        lags (pl.DataFrame | None): Optional lagged data to enhance prediction accuracy.\n\n        Returns:\n        pl.DataFrame | pd.DataFrame: A DataFrame containing predictions with 'row_id' and 'responder_6' columns.\n\n        Raises:\n        TypeError: If the returned predictions are not a DataFrame.\n        \n        Notes:\n        - Aligns column types between cached data and lag data.\n        - Joins lag data with cached test data to fill missing values.\n        - Uses an ARIMA model to generate predictions.\n        - Ensures that the output DataFrame has the same number of rows as the input test data.\n        \"\"\"\n        \n        if lags is not None:\n            # Cast columns to a common data type if necessary\n            self.cached_test_data = self.cached_test_data.with_columns([\n                pl.col(\"date_id\").cast(pl.Int32),\n                pl.col(\"time_id\").cast(pl.Int32),\n                pl.col(\"symbol_id\").cast(pl.Int32)\n            ])\n            \n            lags = lags.with_columns([\n                pl.col(\"date_id\").cast(pl.Int32),\n                pl.col(\"time_id\").cast(pl.Int32),\n                pl.col(\"symbol_id\").cast(pl.Int32)\n            ])\n            \n            lags = lags.with_columns(\n                (pl.col('date_id') - 1).alias('date_id')\n            ).select(['date_id','time_id','symbol_id','responder_6_lag_1'])\n\n            self.cached_test_data = self.cached_test_data.join(\n                lags,\n                on=['date_id','time_id','symbol_id'],\n                how='left'\n            )\n\n            self.cached_test_data = self.cached_test_data.with_columns(\n                pl.when((pl.col(\"responder_6\") == 0.0) & (pl.col('responder_6_lag_1') != 0.0))\n                .then(pl.col('responder_6_lag_1'))\n                .otherwise(pl.col('responder_6'))\n                .alias(\"responder_6\")\n            ).drop(['responder_6_lag_1'])\n            \n\n            trainer = ArimaTrainer(\n                train_size=self.train_size,\n                data=self.cached_test_data.to_pandas(),\n            )\n\n            symbol, pred = trainer.run()\n\n            self.symbol_arr = np.array(symbol)\n            self.pred_arr = np.array(pred)\n\n        # Get unique symbols and their first occurrences\n        _, first_indices = np.unique(self.symbol_arr, return_index=True)\n\n        symbol = self.symbol_arr[first_indices]\n        pred = self.pred_arr[first_indices]\n\n        self.symbol_arr = np.delete(self.symbol_arr, first_indices)\n        self.pred_arr = np.delete(self.pred_arr, first_indices)\n\n        pred_df = pl.DataFrame({\n            'symbol_id': symbol,\n            'responder_6': pred\n        },\n            schema={\n                'symbol_id': pl.Int8,\n                'responder_6': pl.Float32\n            }\n        )\n        \n        self.cached_test_data = pl.concat([\n            self.cached_test_data, \n            test.with_columns(\n                pl.lit(0.0).cast(pl.Float32).alias('responder_6')\n            ).select(['date_id','time_id','symbol_id','responder_6','weight'])\n        ],how='vertical_relaxed')\n\n        predictions = test.join(pred_df, on=['symbol_id'], how='left').select(['row_id','responder_6'])\n    \n        if isinstance(predictions, pl.DataFrame):\n            assert predictions.columns == ['row_id', 'responder_6']\n        elif isinstance(predictions, pd.DataFrame):\n            assert (predictions.columns == ['row_id', 'responder_6']).all()\n        else:\n            raise TypeError('The predict function must return a DataFrame')\n        \n        # Confirm has as many rows as the test data.\n        assert len(predictions) == len(test)\n\n        print(predictions.head())\n    \n        return predictions","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T20:24:26.096409Z","iopub.execute_input":"2024-12-02T20:24:26.096821Z","iopub.status.idle":"2024-12-02T20:24:26.114480Z","shell.execute_reply.started":"2024-12-02T20:24:26.096775Z","shell.execute_reply":"2024-12-02T20:24:26.113456Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def get_last_n_dates_per_symbol(file_path: str, train_size: int) -> pl.DataFrame:\n    \"\"\"\n    Load data from parquet file and return the last n dates for each symbol_id\n    \n    Args:\n        file_path: Path to the parquet file\n        train_size: Number of distinct dates to keep per symbol\n        \n    Returns:\n        pl.DataFrame: Filtered dataframe with last n dates per symbol\n    \"\"\"\n    # Load data\n    data = pl.scan_parquet(file_path)\n    \n    # Get unique dates per symbol\n    unique_dates = (\n        data\n        .select(['symbol_id', 'date_id'])\n        .unique()\n    )\n    \n    # Sort and get top N dates per symbol\n    top_dates = (\n        unique_dates\n        .sort(['symbol_id', 'date_id'], descending=[False, True])\n        .group_by('symbol_id')\n        .head(train_size)\n    )\n    \n    # Filter original data using these dates\n    final_result = (\n        data\n        .join(\n            top_dates,\n            on=['symbol_id', 'date_id'],\n            how='inner'\n        )\n        .sort(['symbol_id', 'date_id', 'time_id'])\n        .select(\n            ['date_id', 'time_id', 'symbol_id', 'responder_6', 'weight']\n        )\n        .collect()\n    )\n    \n    # Validate that each symbol has exactly train_size distinct dates\n    date_counts = final_result.group_by(\"symbol_id\").agg(\n        pl.col(\"date_id\").n_unique().alias(\"distinct_date_count\")\n    )\n    \n    assert (date_counts['distinct_date_count'] == train_size).all(), (\n        f\"Not all symbols have exactly {train_size} distinct dates. \"\n        f\"Found counts: {date_counts.filter(pl.col('distinct_date_count') != train_size)}\"\n    )\n    \n    return final_result","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T20:24:26.257406Z","iopub.execute_input":"2024-12-02T20:24:26.257788Z","iopub.status.idle":"2024-12-02T20:24:26.266034Z","shell.execute_reply.started":"2024-12-02T20:24:26.257756Z","shell.execute_reply":"2024-12-02T20:24:26.264824Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"file_path = '/kaggle/input/jane-street-real-time-market-data-forecasting/train.parquet'\ntrain_size = 21\ndata = get_last_n_dates_per_symbol(file_path, train_size)\n\njane_predictor = JanePredictor(\n    initial_data = data,\n    train_size = train_size\n)\n\ninference_server = kaggle_evaluation.jane_street_inference_server.JSInferenceServer(jane_predictor.predict)\n\nif os.getenv('KAGGLE_IS_COMPETITION_RERUN'):\n    inference_server.serve()\nelse:\n    inference_server.run_local_gateway(\n        (\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/test.parquet',\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/lags.parquet',\n        )\n    )","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2024-12-02T20:24:26.267868Z","iopub.execute_input":"2024-12-02T20:24:26.268241Z","iopub.status.idle":"2024-12-02T20:25:45.797079Z","shell.execute_reply.started":"2024-12-02T20:24:26.268189Z","shell.execute_reply":"2024-12-02T20:25:45.795753Z"}},"outputs":[],"execution_count":null}]}