LIVE

Units, sources, and methods are explicit Deflators & rebasing documented We show revisions and caveats No cherry-picking: full context

Methodology

Our standard rules for transforming raw data into charts and metrics — from units and index bases to smoothing, deflators, and composite scores.

Summary Charting rules Adjustments

At-a-glance

Units & currency

We label units; currency is INR unless stated. USD conversions use daily RBI reference or exchange quotes as noted.

Deflators & bases

Real series use CPI/WPI/GDP deflators. Index bases (100 = base-date) are specified in captions.

Smoothing

We prefer trailing averages (3/6/12-M) with raw series also shown or available.

Axes Axes and units are explicit; left axis for primary measure, right axis used sparingly.
Scales Linear by default; log scale when growth ratios matter or spans are wide (clearly labeled).
Ranges Zero baseline for quantities; trimmed ranges allowed when emphasizing variation (annotated).
Colors High contrast, color-blind friendly. One series = one hue; uncertainty bands use opacity.
Labels Titles state metric, geography, and period; subtitles mention deflators or adjustments.
Data gaps Missing values are not interpolated unless stated; gaps are visually indicated.

Topic	Default	Notes
Rebasing	Index = 100 at base date	Caption: Base=YYYY-MM
Deflation	CPI (headline) or WPI; GDP deflator for macro totals	Choice stated; alt deflators noted when relevant
Seasonal adjust.	Raw by default	SA/SAAR labeled if used; method referenced
Smoothing	3-M trailing average	Raw series available on hover or secondary chart
Growth rates	MoM/YoY as context dictates	Compound/annualized rates labeled “AR”
Currency	INR	USD converted with RBI reference or market close; date shown

Inputs are standardized (z-scores) over a rolling window unless level units are comparable.
Weights are equal unless stated; expert or regression weights are disclosed.
Outliers beyond ±4σ are winsorized or annotated with rationale.
Backfilled history uses consistent methodology; breaks are flagged.

Survey-based estimates show confidence intervals where available.
Model estimates are labeled “Estimate”; bands represent model/past error, not guarantees.
We avoid false precision; round to meaningful granularity.

When sources revise history, we refresh affected charts promptly and annotate material effects.
Base-year or methodology changes are noted in captions and, for major series, in a brief explainer.
We archive prior outputs when comparisons matter (e.g., GDP series across base changes).

Backtests are hypothetical and reflect assumptions shown on the page.
We use realistic costs/slippage where relevant, and state survivorship/data snooping caveats.
Benchmarks and windows are pre-declared; no cherry-picking of favorable periods.

Each chart lists “Source(s)” with links plus “bataSutra calculations” when derived.
Method notes appear in a tooltip/footnote; full details link to this Methodology page.
See Data Sources for our source catalog and licensing notes.

Method notes refreshed Composite weights listed Deflator choices clarified Revisions policy updated Backtest caveats emphasized