Analytics & Modeling | Savvy Analytic Solutions

Featured Interactive Tool

Built on NSOC Regression Analysis

Interactive Calculator

Caregiver Strain ROI Calculator

Quantifies productivity loss by caregiver strain level and calculates employer ROI from caregiver-support benefit investments. Powered by ordinal logistic regression results from the National Study of Caregiving (NSOC), with strain-level odds ratios ranging from 4× to 40×.

40×

High-strain OR

4×

Low-strain OR

1,058

NSOC sample n

p<.001

Model sig.

Launch Calculator ↗ View Research ↗

Caregiver Strain → Productivity Loss Odds Ratios

No Strain

—

1×

Low Strain

4×

Moderate

15×

High Strain

40×

Source: NSOC Ordinal Logistic Regression · n=1,058 · p<.001

Interactive Tools & Dashboards

Live Analytic Applications

Health Analytics · Interactive

Caregiver Work Productivity Risk Stratification Tool

Stratifies employees by caregiver strain level and predicts productivity risk tier. Powered by NSOC ordinal logistic regression — 40× peak odds ratio. Built for HR and benefits decision-makers.

Risk StratificationLogistic RegressionNSOCInteractive

Workforce Analytics · Dashboard

Multigenerational Workforce Benefits Dashboard

Visualizes benefits utilization, engagement, and health priorities segmented by generation (Gen Z, Millennial, Gen X, Boomer). Supports targeted benefits strategy and workforce planning decisions.

MultigenerationalBenefits AnalyticsWorkforceInteractive

Strategic Analytics · OKR

TotalHealth OKR Performance Dashboard — FY2024

Executive-level OKR tracking dashboard linking workforce health objectives to measurable key results. Built for health plan and employer stakeholders to monitor program performance and goal attainment.

OKR TrackingHealth PlanExecutive ReportingInteractive

Statistical Models

Formal Analyses & Applied Research

Survival Analysis · Stata

Discrete-Time Survival Analysis: Young Adult Worker Mental Health

Survey-weighted discrete-time hazard models examining timing of depression and loneliness onset across 9 Household Pulse Survey cycles. Grounded in Job Demands-Resources theory with marginsplot visualizations.

svy: logitxtlogitmarginsplotn=7,000+

Multivariate Regression

Impact of Consumer Call Campaign on Order Volume

Retail POS data regression analysis identifying call volume as a significant negative predictor of order volume (−1.398 units/call, p<.001). Adj. R² = 0.64. Interaction and non-linear terms examined.

Stepwise RegressionVIFAdj R²=0.64

Time Series · SAS

Seasonal Demand Forecasting: ARIMA Model with 12-Period Ahead Forecast

PROC ARIMA with seasonal decomposition and spectral analysis. Auto-selected ARIMA(1,1,1)(1,1,1) specification for operational call volume forecasting. Residual diagnostics and confidence interval output.

PROC ARIMASeasonal12-Period Forecast

Code Showcase

SQL · Stata · SAS · SAS ARIMA

canadian_member_claims_integration.sql

-- Canadian Member Claims Integration Query
-- Integrates member demographics, claims activity, and digital health
-- platform engagement into a unified analytic dataset.

CREATE OR REPLACE TABLE ANALYTICS_DB.canadian_member_claims AS

WITH base_claims AS (
  SELECT
      c.member_id, c.claim_id, c.claim_type, c.diagnosis_group
    , c.service_date, c.paid_amount, c.provider_specialty
    , c.province, c.country
  FROM CLAIMS_DB.claim_header c
  WHERE c.country  =  'Canada'
    AND c.province <> 'QC'
),

platform_activity AS (
  SELECT
      p.member_id
    , COUNT(*)                                                        AS total_platform_events
    , COUNT(DISTINCT p.program_id)                                      AS distinct_programs
    , MAX(p.last_active_date)                                            AS last_platform_activity
    , SUM(CASE WHEN p.event_type = 'COACHING_SESSION' THEN 1 ELSE 0 END) AS coaching_sessions
    , SUM(CASE WHEN p.event_type = 'DIGITAL_CHECKIN'  THEN 1 ELSE 0 END) AS digital_checkins
  FROM PLATFORM_DB.member_activity p
  GROUP BY p.member_id
),

claim_aggregates AS (
  SELECT
      member_id
    , COUNT(DISTINCT claim_id)       AS total_claims
    , SUM(paid_amount)                 AS total_paid
    , AVG(paid_amount)                 AS avg_paid_per_claim
    , COUNT(DISTINCT diagnosis_group)  AS distinct_conditions
    , MIN(service_date)                AS first_claim_date
    , MAX(service_date)                AS last_claim_date
  FROM base_claims
  GROUP BY member_id
)

SELECT
    mp.account_owner, mp.member_id AS cuid
  , mp.billing_province, mp.market_segment, mp.primary_provider
  , mp.first_eligibility_year, mp.last_eligibility_year
  , ca.total_claims, ca.total_paid, ca.avg_paid_per_claim
  , ca.distinct_conditions, ca.first_claim_date, ca.last_claim_date
  , pa.total_platform_events, pa.distinct_programs
  , pa.last_platform_activity, pa.coaching_sessions, pa.digital_checkins
  , CASE WHEN ca.total_claims          > 0 THEN 1 ELSE 0 END AS has_claims
  , CASE WHEN pa.total_platform_events > 0 THEN 1 ELSE 0 END AS engaged_in_platform
FROM MEMBER_DB.member_dim mp
  LEFT JOIN claim_aggregates ca ON mp.member_id = ca.member_id
  LEFT JOIN platform_activity pa ON mp.member_id = pa.member_id
WHERE mp.country = 'Canada' AND mp.billing_province <> 'QC'
ORDER BY mp.market_segment, mp.account_owner, mp.member_id;

// Discrete-Time Survival Analysis: Depression Onset
// Household Pulse Survey · Survey-weighted · JD-R Theory

svyset [pw=pweight]

// Model D4: Full JD-R Model
svy: logit dep_first_event i.cycle ///
    dem_fininst not_married ///
    i.work_arrangement i.social3 ///
    i.age_group i.gender i.race_ethnicity ///
    i.education i.ann_income i.region
estimates store D4_full

// Model D5: Interaction — Work Arrangement × Social Connection
svy: logit dep_first_event i.cycle ///
    dem_fininst not_married ///
    i.work_arrangement##i.social3 ///
    i.age_group i.gender i.race_ethnicity ///
    i.education i.ann_income i.region

testparm i.work_arrangement#i.social3

margins work_arrangement, at(social3=(1 2 3)) ///
    vce(unconditional) post
marginsplot, title("Predicted Depression Hazard", size(small)) ///
    xtitle("Social Connection Frequency") ///
    ytitle("P(Depression Onset | At Risk)") ///
    scheme(stgcolor_mv)
graph export "margins_depression_interaction.png", replace width(1200)

/* Tuition Pricing Model — PROC GLMSELECT */
/* n=1,283 · Stepwise selection · Adj R²=0.71 */

PROC MI DATA=project OUT=proj_impute SEED=97071 NIMPUTE=1;
  fcs;
  VAR tuition top25 sf_ratio fac_comp collrate
      graduat pct_phd fulltime alumni num_enrl parochial;
RUN;

PROC GLMSELECT DATA=proj_impute PLOTS=All SEED=523654;
  CLASS parochial;
  PARTITION fraction(Test=0.4);
  MODEL tuition = top25 sf_ratio fac_comp collrate
      graduat pct_phd fulltime alumni num_enrl parochial
      / Selection=Stepwise(Select=SL SLE=0.15 SLS=0.15
        Choose=AdjRSQ) Details=ALL Hierarchy=Single
        Stats=ALL showpvalues;
  OUTPUT Out=Orig;
RUN;

PROC REG DATA=proj_impute;
  MODEL tuition = sf_ratio fac_comp graduat pct_phd
      fulltime alumni parochial enrlproch / PARTIAL;
RUN;

/* ARIMA Demand Forecasting — 12-Period Ahead */
/* Operations Call Volume · Seasonal ARIMA(1,1,1)(1,1,1) */

PROC ARIMA DATA=monthly_calls;
  IDENTIFY VAR=call_volume(1,12)
    NLAG=36 STATIONARITY=(ADF=2);
RUN;

PROC ARIMA DATA=monthly_calls;
  IDENTIFY VAR=call_volume(1,12);
  ESTIMATE P=(1) Q=(1) SP=(1) SQ=(1) NOSTABLE;
  FORECAST LEAD=12 INTERVAL=month
    ID=date OUT=forecast_out;
RUN;

DATA forecast_final;
  SET forecast_out;
  IF _TYPE_ = 'FORECAST';
  upper_bound = forecast + 1.96 * std;
  lower_bound = forecast - 1.96 * std;
  FORMAT date MONYY7.;
  KEEP date forecast std upper_bound lower_bound;
RUN;

Visualizations & Infographics

Complex models to decision-ready visuals

Caregiver Strain & Workforce Productivity

NSOC regression findings translated into a clear, executive-ready narrative.

NSOC

Dataset

Regression

Work

Impact

Regression Caregiving Productivity

View infographic

Mental Health & Work Arrangement

HPS survival analysis visualized for intuitive interpretation.

HPS

Dataset

Hazard Ratio

Hybrid

Work

Survival Models Mental Health Work Arrangement

View infographic

"I Can Tell the Story in Any Medium"

Power BI, Excel, Canva, and PowerPoint—same analytic spine, different audiences.

Dashboards

PPT

Decks

Canva

Visuals

Power BI Excel Canva PowerPoint

Explore gallery

Full Analytics Lifecycle

Methodology to Action

Descriptive

Data prep, EDA, and baseline profiling

→

Diagnostic

Root cause analysis and pattern detection

→

Predictive

Regression, survival, and forecasting models

→

Prescriptive

ROI modeling and decision frameworks

→

Communicate

Dashboards, reports, and executive summaries

Let's Build Something Together

Available for full-time remote roles and consulting engagements in health analytics, people analytics, and workforce health research.

Start a Conversation → View CV

Analytics & Modeling Portfolio