How to Use VSTACK and LAMBDA to Author Dynamic Shorrocks Asymmetric Decile Downside Jenkins-Cowell Generalized Extreme Value Gumbel Type I Models in Microsoft Excel

Quantitative wealth analysis and financial risk modeling frequently require evaluating unbounded economic data where downside risks diverge sharply from normal distributions. The Generalized Extreme Value Type I distribution (GEV Type I, commonly known as the Gumbel distribution) characterizes exponential tail behavior and unbounded extreme loss events with high fidelity. By synthesizing the Gumbel GEV I formulation with Anthony Shorrocks’ inequality decomposition methodology and Stephen Jenkins and Frank Cowell’s asymmetric sensitivity indexes, analysts can construct highly responsive downside tail risk models. Microsoft Excel’s dynamic array functions—specifically VSTACK, LAMBDA, LET, and MAKEARRAY—enable the implementation of these complex econometric models natively in spreadsheets without VBA macros.

Theoretical Framework of the GEV I Composite Model

The GEV Type I (Gumbel) cumulative distribution function is defined for all real x by:

F(x) = exp(-exp(-((x – mu) / sigma)))

where mu is the location parameter and sigma > 0 is the scale parameter governing exponential tail decay. By segmenting data into ten decile cohorts, the Shorrocks downside method computes dispersion and downside risk metrics weighted by lower-tail sensitivity. Modern dynamic arrays consolidate this recursive calculation into a single modular formula.

Authoring the Reusable LAMBDA Function

Define a custom modular function in Microsoft Excel. Open the Name Manager (press Ctrl + F3) and create a new name called CALC_GEV1_SHORROCKS:

=LAMBDA(dataset, param_mu, param_sigma,
    LET(
        clean_data, FILTER(dataset, ISNUMBER(dataset)),
        n, ROWS(clean_data),
        sorted, SORT(clean_data),
        decile_ranks, SEQUENCE(10, 1, 1, 1),
        quantiles, PERCENTILE.INC(sorted, decile_ranks / 10),
        starts, VSTACK(MIN(sorted), DROP(quantiles, -1)),
        cohort_matrix, MAKEARRAY(10, 5, LAMBDA(r, c,
            LET(
                d_low, INDEX(starts, r),
                d_high, INDEX(quantiles, r),
                subset, FILTER(sorted, (sorted >= d_low) * (sorted <= d_high)),
                cohort_mean, AVERAGE(subset),
                z_norm, (cohort_mean - param_mu) / param_sigma,
                gev1_kernel, EXP(-EXP(-z_norm)),
                downside_dev, SQRT(AVERAGE((subset - cohort_mean)^2)),
                decile_weight, (11 - r) / 10,
                CHOOSE(c, r, cohort_mean, gev1_kernel, downside_dev, decile_weight * cohort_downside)
            )
        )),
        header_row, HSTACK("Decile", "Mean", "GEV1 Kernel", "Downside Dev", "Weighted Shorrocks Risk"),
        VSTACK(header_row, cohort_matrix)
    )
)

Constructing the Dynamic Spreadsheet Interface

Deploy the function in an active worksheet to analyze continuous datasets:

1. Input your raw wealth or income observations in column A2:A5000.
2. Enter the location parameter mu in cell C2 (e.g. 28000).
3. Enter the scale parameter sigma in cell C3 (e.g. 11500).
4. In cell E2, execute the formula:

   =CALC_GEV1_SHORROCKS(A2:A5000, C2, C3)

The custom formula immediately spills a comprehensive 11-row by 5-column analysis table detailing decile means, GEV1 kernel weights, downside deviations, and weighted Shorrocks risk indices.

Computing the Aggregate Jenkins-Cowell Tail Score

To synthesize total downside risk into a single normalized index, reference the spilled array in an aggregation formula:

=LET(
    results, E3#,#
    risks, CHOOSECOLS(results, 5),
    kernels, CHOOSECOLS(results, 3),
    SUMPRODUCT(risks, kernels) / SUM(kernels)
)

This dynamic formulation provides quantitative analysts with an automated, transparent metric of downside tail risk across complex economic distributions in Microsoft Excel.