
DS-GA 3001 - Fall 2026
NYU Center for Data Science
2026-11-17
Parts 1 and 2 are based on materials by Enrico Bertini (NYU Tandon).
Data in which values depend on time, and time is explicitly recorded.
Time + object (with attributes)
“Something happened at time T”
Time + measure(s)
“This was the value at time T”
Counting events per interval turns event data into measurement data.
Most real series have all three at once. You choose which one a chart emphasizes.
Sequential: sales by month, 2013 to 2016

Cyclic: number of orders by day of the week

Neither chart can show what the other shows.
What aggregation buys
Daily to monthly averages makes a seasonal pattern readable.
What aggregation costs
A one-hour outage is invisible in a monthly average.
Choose the granularity from the question: trends want coarse bins, anomalies want fine ones.
Position encodes both time and value; the line’s slope encodes the rate of change.
The aspect ratio (width / height) changes how steep every change looks.


Cleveland, W. S., McGill, M. E., & McGill, R. (1988). The shape parameter of a two-variable graph. JASA.
Three categories on shared axes are easy to compare.
Seventeen categories on shared axes: color runs out and lines hide each other.
Filter: show only the series you care about

Highlight: keep the rest as faded context

Four regions as lines

The same four regions stacked

Only the bottom layer and the total have a flat baseline. Every other layer rides on the ones below it.
Color replaces height, so many series fit in the space of one line chart.
DeBold, T., & Friedman, D. (2015). Battling Infectious Diseases in the 20th Century: The Impact of Vaccines. The Wall Street Journal.

Line charts squeezed into thin rows

The same rows as horizon charts

Same data, same space. The next two slides show how the right-hand version is built.
Heer, J., Kong, N., & Agrawala, M. (2009). Sizing the horizon: the effects of chart size and layering on the graphical perception of time series visualizations. CHI.


Fill the area between the line and the baseline, then slice it into bands of equal height. Darker means further from the baseline.


Mirror the negative values above the baseline, then slide every band down to the baseline. Height becomes color intensity.
“Small, intense, word-sized graphics with typographic resolution.” (Edward Tufte, Beautiful Evidence, 2006)

No axes and no labels: a sparkline sits beside a number and shows where it came from.
What we ask a model to do
What time changes
The evaluation tools from Week 3 assumed none of these three. Each one has a visual check.

Synthetic daily demand. Linear model on a time index, a yearly cycle and day of week; band from the model’s errors on its last 180 training days.

Same synthetic series; gradient-boosted trees on time index, day of week and day of year.

Same synthetic series and the linear model from two slides back; six folds, each tested on the next 120 days.

Same synthetic series; the linear model with the day-of-week inputs removed.

Same synthetic series; a separate linear model per horizon, each given the 14 most recent days observed when the forecast is issued. All horizons are scored on the same 93 days, each forecast issued on or after the last training day.
A model learns \(P(y \mid x)\) from data drawn from \(P(x)\). After deployment either can change.
What we can watch
Gama, J., Žliobaitė, I., Bifet, A., Pechenizkiy, M., & Bouchachia, A. (2014). A survey on concept drift adaptation. ACM Computing Surveys.

The shape decides how to respond: a sudden change calls for retraining, a recurring one for keeping the old model around.
Schematic, after Gama et al. (2014).

Synthetic stream: a two-feature classifier scored on 400 new labeled points each week. The labeling rule changes at week 44.

Same synthetic stream. Weeks 20 to 32: the inputs shift and the labeling rule is unchanged. Week 44: the rule changes and the inputs do not.

Schematic series. Dotted lines mark the highest and lowest values of the normal series.
Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: a survey. ACM Computing Surveys.

Synthetic demand series, last 240 days, with 12 injected anomalies; σ is a robust estimate of the forecast’s typical error on the training data. A retrospective demonstration: the first 120 of these days are days the forecast model was trained on.

Same detector and data as the previous slide; the dashed line marks k = 3.