Metric Summary Cards#

These cards give a compact readout of the selected series and place the latest value into statistical context.

Current value#

x_{t}

(1)

The latest available data point of the selected series.

Values are displayed in compact form such as K, M, B, and T when appropriate, while still preserving meaningful decimals.

Percentile (4Y) and Zone#

{P e r c e n t i l e}_{4 Y} (x_{t})

(2)

Percentile shows where the current value sits within the last 4 years of daily data, using up to 1460 observations.

If the series is shorter than 4 years, all available history is used instead. If fewer than 90 observations are available, the percentile is hidden.

Zone groups that percentile into broad ranges for faster interpretation:

Cold — below the 10th percentile
Cool — 10th to 25th percentile
Neutral — 25th to 75th percentile
Warm — 75th to 90th percentile
Hot — 90th percentile and above

7D change#

Δ_{7 D} (%) = \frac{x_{t} - x_{t - 7}}{x_{t - 7}} \times 100

(3)

90D change#

Δ_{90 D} (%) = \frac{x_{t} - x_{t - 90}}{x_{t - 90}} \times 100

(4)

The line reading from ... (90D ago) shows the actual value from 90 days earlier:

x_{t - 90}

(5)

Distance#

Shows how far the current value is from the extremes of the same 4-year window used for the percentile. If 4 years of data are not available, full history is used instead.

To 4Y High#

{T o H i g h}_{4 Y} (%) = \frac{x_{t} - max (x)}{| max (x) |} \times 100

(6)

Example: −67% to 4Y High means the current value is 67% below the 4-year high.

From 4Y Low#

{F r o m L o w}_{4 Y} (%) = \frac{x_{t} - min (x)}{| min (x) |} \times 100

(7)

Example: +414% from 4Y Low means the current value is 414% above the 4-year low.

z-Score (robust)#

z_{r o b u s t, 365 D} = \frac{x_{t} - {m e d i a n}_{365 D}}{1.4826 \times {M A D}_{365 D}}

(8)

A 365-day robust z-score based on the median and MAD, or Median Absolute Deviation.

The factor 1.4826 rescales MAD so it is comparable to standard deviation under a normal distribution.

The label, such as Normal, Elevated, or Extreme, follows common cutoffs around ±1σ and ±2σ.

Volatility#

σ_{30 D, a n n} = s t d e v (\ln (\frac{v_{t}}{v_{t - 1}})) \times \sqrt{365}

(9)

30-day annualized volatility based on log returns.

This value is only computed for strictly positive series.

Notes#

All windows assume daily data.

Missing or non-finite values are skipped. If the series is shorter than a given window, all available data is used instead.