thesisPrototype
dataprediction pack · v0.1.0

Consumer-spending nowcast

Combines retail control-group momentum, auto sales, gasoline receipts, and card-spending signals before spending releases.

1
runs
1
targets
1
agents
1
versions

Purpose

Separate true consumer-demand momentum from volatile retail-release channels before advance sales and PCE spending releases.

Mechanism

  • Start from recent headline and control-group retail sales momentum.
  • Split autos, gasoline, food services, and control-group channels before recombining.
  • Cross-check the channel signal against card-spending and price-level nowcasts when available.

Checks

  • The run distinguishes headline retail sales from control-group demand.
  • Auto and gasoline channels are treated separately from broad consumer demand.
  • The interval reflects advance-release volatility and later revision risk.

Inputs

  • Census MARTS advance retail sales history
  • Auto-sales and gasoline-price signals
  • Card-spending and control-group nowcasts

Limitations

  • Prototype card-spending inputs are qualitative rather than a direct feed.
  • Does not yet model seasonal retail-calendar effects at the store-category level.

Versions

v0.1.0consumer-spending-nowcast@0.1.01 runs

Runs using this pack

May retail sales pack set
Retail sales, May 2026 month-over-month
brier-1.packedJun 15, 2026Brier-1 - spending packs
pack_set.retail_may_2026.v1pack-enabled
+0.3%
+0.1% vs no-pack

Pack-enabled run using consumer-spending nowcast and release-vintage checks.

80% interval -0.3% to 0.8%