`Ffill`

Description

The Ffill class performs forward filling on a sequence of data, replacing any NaN values with the most recent non-NaN value. This approach is commonly used to handle missing values by carrying forward the last valid observation until a new valid value appears in the data.

Parameters: Ffill takes no parameters; it simply operates over a data sequence and forward fills any NaN values encountered.

NaN handling: If a NaN appears at the start of the data sequence, it will remain as NaN because no preceding value exists to carry forward.

Usage Example and Plot

Below is an example of using Ffill to forward-fill NaN values in a data sequence, along with a plot showing the original and filled data.

import numpy as np
import plotly.graph_objects as go
from plotly.subplots import make_subplots
from screamer import Ffill

# Generate example data with NaN values
data = np.random.normal(size=30)
data[[3, 6, 10, 15, 21]] = np.nan  # Introduce NaN values

# Apply forward fill
ffilled_data = Ffill()(data)

# Create subplots with specified row heights and shared x-axis
fig = make_subplots(
    rows=2, cols=1,
    shared_xaxes=True,
    row_heights=[1/2, 1/2],
    vertical_spacing=0.1
)

# Add traces for original data and forward-filled data
fig.add_trace(go.Scatter(y=data, mode='lines+markers', name='Original Data'), row=1, col=1)
fig.add_trace(go.Scatter(y=ffilled_data, mode='lines+markers', name='Forward-Filled Data', line=dict(color='red')), row=2, col=1)

# Update layout with individual y-axis titles
fig.update_layout(
    title="Forward Fill (Ffill) on Data with NaNs",
    xaxis_title="Index",
    yaxis_title="Original Data",
    yaxis2_title="Forward-Filled Data",
    margin=dict(l=20, r=20, t=80, b=20),
    legend=dict(orientation="h", yanchor="bottom", y=1.02, xanchor="right", x=1)
)

fig.show()

Implementation Details

Algorithm for `Ffill`

The Ffill function iterates over the data, replacing each NaN value with the most recent valid observation, if available. This ensures that each missing value is filled in a forward direction, but any NaN values at the beginning remain unchanged due to the absence of a prior value.

Complexity

Time Complexity: O(1).
Space Complexity: O(1).

Performance

Ffill is a lightweight operation that process data efficiently. They are suitable for real-time or streaming data applications where missing values need to be managed with minimal overhead.

Ffill