Demand Curve Estimation Workbook¶
Estimation workbook — instantiates Demand Curve Calibration and Response Design
The auditable ledger that assembles every observed cost-quantity-segment observation into a single, uncertainty-tagged demand schedule.
A Demand Curve Estimation Workbook is the plain, disciplined ledger in which a demand schedule is actually assembled and kept. Each row ties an observed quantity to the exact cost conditions that produced it — the price, the wait, the eligibility rule, the time window — along with the segment it came from, the source of the number, and a confidence note. Fitted or interpolated, those rows become a schedule of quantity against generalized cost, with the range over which it is trustworthy marked on its face. Its defining virtue is not clever estimation but bookkeeping integrity: it is the record of record, the single place where every point is traceable to its conditions and no observation floats free of the cost that generated it.
Example¶
A national park is setting camping fees across its reservable sites and needs one honest schedule rather than a folklore of "the north loop always sells out." Staff open a workbook and enter, row by row, each season's data: fee level, nights booked, the booking window that was open, whether it was a holiday weekend, and the source (the reservation system vs. a hand count of walk-ups). Some rows carry fat error bars — the year the reservation site changed vendors is flagged as low-confidence — and rows where every night sold out are marked as censored, because the true quantity sought was hidden by capacity.
From the clean rows the workbook fits a modest schedule: bookings fall gently as fees rise up to about $30 a night, then more steeply past $35, with the reliable range explicitly capped at $40 because nothing above that was ever observed. The output is unglamorous — a table, a fitted line, and a shaded band saying "do not read this curve past $40." But it is the artifact every later decision points back to, and when a manager proposes a $55 tier, the workbook's own extrapolation band is what says we have no evidence there.
How it works¶
- Define the unit first. Fix what one row's "quantity" means — a booked night? a unique party? a completed stay? — before any number goes in, so the column is comparable down its length.
- Enter observations with their conditions. Every quantity is logged next to the full cost setting and segment that produced it, plus provenance and a confidence tag.
- Fit within the evidence. Interpolate or fit a simple schedule across the observed cost range; do not draw the line where there are no points.
- Mark the valid band. Shade the cost region the estimate is trustworthy in and label everything outside it as extrapolation.
Tuning parameters¶
- Quantity definition — attempted vs. completed vs. rationed demand; the choice sets what the whole schedule is a curve of, and mixing them corrupts it.
- Aggregation level — one pooled curve is stable but blurs segments; splitting rows by segment reveals differences at the cost of thinner, noisier cells.
- Fit stiffness — a rigid functional form resists overfitting noise but can miss a real kink; a flexible fit tracks the points but chases scatter.
- Confidence coding — how strictly low-quality rows are down-weighted or excluded; strictness buys trust but discards information.
- Extrapolation cap — how far past the observed range the schedule is allowed to be read at all.
When it helps, and when it misleads¶
Its strength is traceability under scrutiny: every point survives an audit, the valid range is explicit, and the schedule can be rebuilt as new rows arrive. It is the artifact that keeps a demand conversation anchored to evidence instead of anecdote.
Its failure mode is quiet: a workbook full of tidy numbers looks authoritative even when the rows are thin, stale, or contaminated, and a fitted line invites reading straight through the gaps. Twyman's law is the standing caution — any figure that looks especially clean or striking is worth doubting, because it is more often an artifact of how the data was recorded than a real feature of demand.[n1] The classic misuse is extrapolating the fitted curve past the shaded band to justify a price never observed. The guarding discipline is to treat the workbook as a record, not an oracle: keep provenance on every row, honor the extrapolation cap, and re-open the book when conditions change rather than trusting a frozen fit.
How it implements the components¶
demand_object_boundary— the mandatory first step fixes the unit and bundle each row counts, so the schedule measures one comparable thing.quantity_sought_measure— each row's quantity is explicitly typed (attempted, completed, censored), keeping the measure honest.demand_schedule_model— the fitted table of quantity against generalized cost is the workbook's central output.uncertainty_and_extrapolation_band— confidence tags and the marked valid range travel with the schedule so its limits are visible.
It does not manufacture new causal evidence — it neither runs the manipulation that estimates a slope (elasticity_and_threshold_profile, observed_response_evidence_base), which is the Price Sensitivity Experiment, nor recovers demand hidden by scarcity (latent_and_suppressed_demand_probe), which is the Waitlist and Stockout Analysis; it records and assembles what those produce.
Related¶
- Instantiates: Demand Curve Calibration and Response Design — the workbook is where the governed schedule physically lives.
- Consumes: Price Sensitivity Experiment and Revealed Preference Choice Log supply the observations its rows record.
- Sibling mechanisms: Conjoint or Discrete Choice Model · Cross-Elasticity Matrix · Demand Segmentation Dashboard · Equity Access Impact Review · Price Sensitivity Experiment · Revealed Preference Choice Log · Scenario Demand Stress Test · Shadow Price Probe · Waitlist and Stockout Analysis
Editorial Notes¶
Form Classification¶
Form family: Analysis, Modeling & Optimization
Rationale: The workbook assembles cost, quantity, segment, provenance, and confidence observations and fits a demand schedule only within the evidenced range, so its operative form is demand estimation.
Nearest alternative: Record, Log & Register — Observed points accumulate audibly in a ledger, but fitting the uncertainty-bounded schedule rather than preserving history alone is the defining purpose.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Economics & Finance
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Economics cohered estimation of quantity demanded against price and other generalized costs, stratified by market segment.
Related originating lineages:
- Statistics & Experimental Design — Statistical estimation supplied uncertainty, interpolation, provenance, and validity-range discipline.
Review resolution: Economics cohered estimation of quantity demanded against price and other generalized costs, stratified by market segment. Economic demand schedules and statistical estimation jointly constitute the auditable workbook, so the synthesis and ambiguity are retained.
Attribution caveat: The demand schedule is economic; the auditable workbook is an encyclopedia implementation artifact.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
[n1] Twyman's law — "any figure that looks interesting or different is usually wrong." A survey-research maxim warning that unusually clean, large, or convenient numbers most often reflect a measurement or recording artifact rather than a real effect, and deserve extra scrutiny before they are trusted. ↩