py_alpaca_api.stock.predictor

Attributes

yesterday

four_years_ago

logger

Classes

History

Screener

Predictor

Module Contents

class py_alpaca_api.stock.predictor.History(data_url: str, headers: dict[str, str], asset: py_alpaca_api.stock.assets.Assets)[source]
BATCH_SIZE = 200
check_if_stock(symbol: str) → py_alpaca_api.models.asset_model.AssetModel[source]

Check if the asset corresponding to the symbol is a stock.

Parameters:

symbol (str) – The symbol of the asset to be checked.

Returns:

The asset information for the given symbol.

Return type:

AssetModel

Raises:

ValueError – If there is an error getting the asset information or if the asset is not a stock.

get_stock_data(symbol: str | list[str], start: str, end: str, timeframe: str = '1d', feed: str = 'sip', currency: str = 'USD', limit: int = 1000, sort: str = 'asc', adjustment: str = 'raw') → pandas.DataFrame[source]

Retrieves historical stock data for one or more symbols within a specified date range and timeframe.

Parameters:

  • symbol – The stock symbol(s) to fetch data for. Can be a single symbol string or list of symbols.

  • start – The start date for historical data in the format “YYYY-MM-DD”.

  • end – The end date for historical data in the format “YYYY-MM-DD”.

  • timeframe – The timeframe for the historical data. Default is “1d”.

  • feed – The data feed source. Default is “sip”.

  • currency – The currency for historical data. Default is “USD”.

  • limit – The number of data points to fetch per symbol. Default is 1000.

  • sort – The sort order for the data. Default is “asc”.

  • adjustment – The adjustment for historical data. Default is “raw”.

Returns:

A pandas DataFrame containing the historical stock data for the given symbol(s) and time range.

Raises:

ValueError – If the given timeframe is not one of the allowed values.

static preprocess_multi_data(symbols_data: dict[str, list[collections.defaultdict]]) → pandas.DataFrame[source]

Preprocess data for multiple symbols.

Parameters:

symbols_data – A dictionary mapping symbols to their bar data.

Returns:

A pandas DataFrame containing the preprocessed historical stock data for all symbols.

static preprocess_data(symbol_data: list[collections.defaultdict], symbol: str) → pandas.DataFrame[source]

Prepross data Preprocesses the given symbol data by converting it to a pandas DataFrame and performing various data transformations.

Parameters:

  • symbol_data – A list of defaultdict objects representing the JSON response data.

  • symbol – A string representing the symbol or ticker for the stock data.

Returns:

A pandas DataFrame containing the preprocessed historical stock data.

get_historical_data(symbols: list[str], url: str, params: dict, is_single: bool) → dict[str, list[collections.defaultdict]][source]

Retrieves historical data for given symbol(s).

Parameters:

  • symbols – List of symbols for which to retrieve historical data.

  • url – The URL to send the request to.

  • params – Additional parameters to include in the request.

  • is_single – Whether this is a single-symbol request.

Returns:

A dictionary mapping symbols to their historical data.

Return type:

dict[str, list[defaultdict]]

get_latest_bars(symbols: str | list[str], feed: str = 'iex', currency: str = 'USD') → pandas.DataFrame | dict[str, pandas.DataFrame][source]

Get the latest bars for one or more symbols.

The latest bars endpoint returns the most recent minute bar for each requested symbol.

Parameters:

  • symbols – Symbol(s) to get latest bars for. Can be a string for single symbol or list of strings for multiple symbols.

  • feed – The data feed to use (“iex”, “sip”, or “otc”). Defaults to “iex”.

  • currency – The currency for the returned prices. Defaults to “USD”.

Returns:

pd.DataFrame with the latest bar data. For multiple symbols: dict mapping symbols to DataFrames with latest bar data.

Return type:

For single symbol

Raises:

  • ValueError – If feed is invalid or symbols is empty.

  • Exception – If the API request fails or returns no data.

class py_alpaca_api.stock.predictor.Screener(data_url: str, headers: dict[str, str], asset: py_alpaca_api.stock.assets.Assets, market: py_alpaca_api.trading.market.Market)[source]
filter_stocks(price_greater_than: float, change_condition: collections.abc.Callable[[pandas.DataFrame], pandas.Series], volume_greater_than: int, trade_count_greater_than: int, total_returned: int, ascending_order: bool) → pandas.DataFrame[source]

Filter stocks based on given parameters.

Parameters:

  • price_greater_than – The minimum price threshold for the stocks.

  • change_condition – A callable function that takes in a DataFrame and returns a boolean Series. This function is used to filter the stocks based on a specific change condition.

  • volume_greater_than – The minimum volume threshold for the stocks.

  • trade_count_greater_than – The minimum trade count threshold for the stocks.

  • total_returned – The number of stocks to return.

  • ascending_order – A boolean value indicating whether to sort the stocks in ascending order by change value.

Returns:

A pandas DataFrame containing the filtered stocks.

losers(price_greater_than: float = 5.0, change_less_than: float = -2.0, volume_greater_than: int = 20000, trade_count_greater_than: int = 2000, total_losers_returned: int = 100) → pandas.DataFrame[source]

Returns a filtered DataFrame of stocks that meet the specified conditions for losers.

Parameters:

  • price_greater_than (float) – The minimum price threshold for stocks to be considered losers. Default is 5.0.

  • change_less_than (float) – The maximum change threshold for stocks to be considered losers. Default is -2.0.

  • volume_greater_than (int) – The minimum volume threshold for stocks to be considered losers. Default is

  • 20000.

  • trade_count_greater_than (int) – The minimum trade count threshold for stocks to be considered losers. Default is 2000.

  • total_losers_returned (int) – The maximum number of losers to be returned. Default is 100.

Returns:

A filtered DataFrame containing stocks that meet the specified conditions for losers.

Return type:

pd.DataFrame

gainers(price_greater_than: float = 5.0, change_greater_than: float = 2.0, volume_greater_than: int = 20000, trade_count_greater_than: int = 2000, total_gainers_returned: int = 100) → pandas.DataFrame[source]
Parameters:

  • price_greater_than (float) – The minimum price threshold for the stocks to be included in the gainers list.

  • 5.0. (Default is)

  • change_greater_than (float) – The minimum change (in percentage) threshold for the stocks to be included in

  • list. (the gainers)

  • 2.0. (Default is)

  • volume_greater_than (int) – The minimum volume threshold for the stocks to be included in the gainers list. Default is 20000.

  • trade_count_greater_than (int) – The minimum trade count threshold for the stocks to be included in the

  • 2000. (gainers list. Default is)

  • total_gainers_returned (int) – The maximum number of gainers to be returned. Default is 100.

Returns:

A Pandas DataFrame containing the stocks that satisfy the criteria for being gainers.

Return type:

pd.DataFrame

set_dates()[source]

Sets the dates for the screener.

This method retrieves the last two trading dates from the market calendar and assigns them to the yesterday and day_before_yesterday attributes.

Returns:

None

py_alpaca_api.stock.predictor.yesterday[source]
py_alpaca_api.stock.predictor.four_years_ago[source]
py_alpaca_api.stock.predictor.logger[source]
class py_alpaca_api.stock.predictor.Predictor(history: py_alpaca_api.stock.history.History, screener: py_alpaca_api.stock.screener.Screener)[source]
get_stock_data(symbol: str, timeframe: str = '1d', start: str = four_years_ago, end: str = yesterday) → pandas.DataFrame[source]

Retrieves historical stock data for a given symbol within a specified timeframe.

Parameters:

  • symbol (str) – The stock symbol to retrieve data for.

  • timeframe (str, optional) – The timeframe for the data. Defaults to “1d”.

  • start (str, optional) – The start date for the data. Defaults to four_years_ago.

  • end (str, optional) – The end date for the data. Defaults to yesterday.

Returns:

A DataFrame containing the historical stock data with columns “ds” (date) and “y” (vwap).

Return type:

pd.DataFrame

static train_prophet_model(data)[source]

Trains a Prophet model using the provided data.

Parameters:

data – The input data used for training the model.

Returns:

The trained Prophet model.

static generate_forecast(model, future_periods=14)[source]

Generates a forecast using the specified model for a given number of future periods.

Parameters:

  • model – The model used for forecasting.

  • future_periods – The number of future periods to forecast.

Returns:

The forecasted value for the next two weeks.

get_losers_to_gainers(gain_ratio: float = 10.0, losers_to_scan: int = 200, future_periods: int = 5) → list[source]