Use Case: Time Series Analysis

LOWESS for trend extraction and temporal smoothing.

Overview

Time series data often contains noise, seasonality, and trends. LOWESS provides flexible trend extraction without parametric assumptions.

Basic Trend Extraction

fraction = 0.1 sizes the neighbourhood as 10% of the data at each evaluation point — narrow enough to follow a slowly varying trend without smearing periodic variation. Three robustness iterations down-weight noise spikes so they cannot bias the fitted curve; this is especially important when the signal-to-noise ratio is low or when occasional outliers are expected.

library(rfastlowess)

n <- 500
t <- seq(0, 100, length.out = n)
i <- 0:(n - 1)
y <- 10 + 0.5 * t + 3 * sin(t / 10) + ((i * 7 + 3) %% 1.7 - 0.85) * 3

# t and y are your time series vectors
model <- Lowess(
    fraction = 0.1,
    iterations = 3
)
result <- fit(model, t, y)

cat(sprintf("y[0]: %.4f\n", result$y[1]))
#> y[0]: 11.3116

Detrending

Remove trend to analyze residual patterns.

Setting outputs = "residuals" stores observed - smoothed alongside the smooth. A slightly wider fraction = 0.3 produces a smoother baseline trend, so short-duration oscillations end up in the residuals rather than being absorbed into the trend component. The residual series is then ready for spectral analysis, seasonality detection, or change-point methods.

library(rfastlowess)

n <- 100
t <- seq(0, 2 * pi, length.out = n)
y <- sin(t) + 0.1

model <- Lowess(
    fraction = 0.3,
    iterations = 3,
    outputs = "residuals"
)
result <- fit(model, t, y)

trend <- result$y
detrended <- result$residuals
cat(sprintf("Trend y[0]: %.4f  residual: %.4f\n", trend[1], detrended[1]))
#> Trend y[0]: 0.2582  residual: -0.1582

Forecasting with Prediction Intervals

Prediction intervals widen the uncertainty band to include both the uncertainty in the fitted curve (confidence interval) and the expected scatter of new observations around it. fraction = 0.2 offers a balance between local detail and stable interval width — too small a fraction produces jagged interval edges; too large a fraction underestimates local variance near turning points.

library(rfastlowess)

n <- 100
t <- seq(0, 2 * pi, length.out = n)
y <- sin(t) + 0.1

model <- Lowess(
    fraction = 0.2,
    iterations = 3,
    intervals = intervals_opts(confidence = 0.90, prediction = 0.99)
)
result <- fit(model, t, y)

cat(sprintf(
    "95%% PI: [%.4f, %.4f]\n",
    result$prediction_lower[1], result$prediction_upper[1]
))
#> 95% PI: [0.1337, 0.3213]

Handling Missing Data

LOWESS naturally handles irregular time sampling:

library(rfastlowess)

t_irregular <- vapply(
    0:99, function(i) i * 1.0 + (i * 31 %% 10) * 0.1, numeric(1)
)
y_irregular <- 10 + 0.3 * t_irregular + 2.0 * sin(t_irregular * 0.1)

# No special handling needed for irregular spacing
model <- Lowess(fraction = 0.2)
result <- fit(model, t_irregular, y_irregular)
cat(sprintf("y[0]: %.4f\n", result$y[1]))
#> y[0]: 11.4517

Multi-Scale Analysis

Use different fractions to extract features at different scales:

library(rfastlowess)

n <- 100
t <- seq(0, 2 * pi, length.out = n)
y <- sin(t) + 0.1

scales <- c(0.05, 0.2, 0.5)
trends <- lapply(scales, function(f) {
    model <- Lowess(fraction = f)
    fit(model, t, y)$y
})
cat("Trend y[0] values:", vapply(trends, function(tr) tr[1], numeric(1)), "\n")
#> Trend y[0] values: 0.131712 0.2275244 0.3325414

Gene Expression Time Course

Biological application:

library(rfastlowess)

hours <- seq(0, 24, by = 0.5)
i <- seq_along(hours) - 1
expression <- 100 * (1 + 0.5 * sin(hours * pi / 12)) +
    ((i * 7 + 3) %% 1.7 - 0.85) * 10

model <- Lowess(
    fraction = 0.3,
    iterations = 3,
    outputs = "diagnostics",
    intervals = intervals_opts(confidence = 0.95)
)
result <- fit(model, hours, expression)
cat(sprintf("R2: %.3f\n", result$diagnostics$r_squared))
#> R2: 0.976

Choosing Fraction for Time Series

Data Type	Recommended Fraction	Rationale
Daily data (years)	0.3–0.5	Capture annual trends
Hourly data (days)	0.1–0.2	Capture daily patterns
Sensor data (minutes)	0.05–0.1	Preserve short-term features
Noisy data	Higher	Reduce noise impact
Clean data	Lower	Preserve detail

sessionInfo()
#> R version 4.6.1 (2026-06-24)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 26.04.1 LTS
#> 
#> Matrix products: default
#> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
#> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.32.so;  LAPACK version 3.12.0
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
#>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
#>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
#> 
#> time zone: Etc/UTC
#> tzcode source: system (glibc)
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] rfastlowess_5.0.0 rmarkdown_2.32   
#> 
#> loaded via a namespace (and not attached):
#>  [1] digest_0.6.39    R6_2.6.1         fastmap_1.2.0    xfun_0.61       
#>  [5] maketools_1.3.2  cachem_1.1.0     knitr_1.52       htmltools_0.5.9 
#>  [9] buildtools_1.0.0 lifecycle_1.0.5  cli_3.6.6        sass_0.4.10     
#> [13] jquerylib_0.1.4  compiler_4.6.1   sys_3.4.3        tools_4.6.1     
#> [17] evaluate_1.0.5   bslib_0.12.0     yaml_2.3.12      otel_0.2.0      
#> [21] jsonlite_2.0.0   rlang_1.3.0