| from typing import Callable |
| |
| import numpy as np |
| |
| |
| def generate_random_walk_time_series( |
| num_datapoints: int, |
| start_value: float, |
| step_mean: float, |
| step_stddev: float, |
| cumulative: bool = True, |
| min_value: float = None, |
| max_value: float = None, |
| apply: Callable = lambda x: x, |
| ) -> np.array: |
| random_nums = np.random.normal(step_mean, step_stddev, num_datapoints) |
| out = [] |
| curr = start_value |
| for random_num in random_nums: |
| if cumulative: |
| curr += random_num |
| else: |
| curr = random_num |
| if min_value is not None: |
| curr = max(curr, min_value) |
| if max_value is not None: |
| curr = min(curr, max_value) |
| out.append(curr) |
| return [apply(item) for item in out] |
