BasePrimary¶

class pfhedge.instruments.BasePrimary[source]¶

Base class for all primary instruments.

A primary instrument is a basic financial instrument which is traded on a market and therefore the price is accessible as the market price. Examples include stocks, bonds, commodities, and currencies.

Derivatives are issued based on primary instruments (See BaseDerivative for details).

Buffers:

spot (torch.Tensor): The spot price of the instrument.

dtype¶

The dtype with which the simulated time-series are represented.

Type: torch.dtype

device¶

The device where the simulated time-series are.

Type: torch.device

buffers()[source]¶

Returns an iterator over module buffers.

Yields: torch.Tensor – module buffer

cpu()¶

Moves all buffers of this instrument and its underlier to the CPU.

Note

This method modifies the instrument in-place.

Returns: self

cuda(device=None)¶

Moves all buffers of this instrument and its underlier to the GPU.

Note

This method modifies the instrument in-place.

Parameters: device (int, optional) – If specified, all buffers will be copied to that device.
Returns: self

property default_init_state¶: Returns the default initial state of simulation.

double()¶

Casts all floating point parameters and buffers to torch.float64 datatype.

Note

This method modifies the instrument in-place.

Returns: self

float()¶

Casts all floating point parameters and buffers to torch.float32 datatype.

Note

This method modifies the instrument in-place.

Returns: self

get_buffer(name)[source]¶

Returns the buffer given by target if it exists, otherwise throws an error.

Parameters: name (str) – the name of the buffer.
Returns: torch.Tensor

half()¶

Casts all floating point parameters and buffers to torch.float16 datatype.

Note

This method modifies the instrument in-place.

Returns: self

named_buffers()[source]¶

Returns an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.

Yields: (string, torch.Tensor) – Tuple containing the name and buffer

register_buffer(name, tensor)[source]¶

Adds a buffer to the instrument. The dtype and device of the buffer are the instrument’s dtype and device.

Buffers can be accessed as attributes using given names.

Parameters

name (str) – name of the buffer. The buffer can be accessed from this module using the given name.
tensor (Tensor or None) – buffer to be registered. If None, then operations that run on buffers, such as cuda, are ignored.

abstract simulate(n_paths, time_horizon, init_state=None)[source]¶

Simulate time series associated with the instrument and add them as buffers.

The shapes of the registered buffers should be (n_paths, n_steps) where n_steps is the minimum integer that satisfies n_steps * self.dt >= time_horizon.

Parameters

n_paths (int) – The number of paths to simulate.
time_horizon (float) – The period of time to simulate the price.
init_state (tuple[torch.Tensor | float], optional) – The initial state of the instrument. If None (default), it uses the default value (See default_init_state).

to(*args, **kwargs)[source]¶

Moves and/or casts the buffers of the instrument.

This can be called as

to(device=None, dtype=None)[source]¶

to(tensor)[source]¶

to(instrument)[source]¶

Its signature is similar to torch.nn.Module.to(). It only accepts floating point dtypes. See Instrument dtype and device for details.

Note

This method modifies the instrument in-place.