Chapter 10 of 17

Gold research committee: from evidence to a risk memo

Turn a bullish or bearish story into a documented research decision with scenarios, counterevidence and implementation gates.

3 min read · Original educational chapter

In this chapter

Write a falsifiable question
Design the evidence packet
Use scenarios before probabilities
Evaluate implementation and the no-position alternative
Require a decision record and a later review
Worked example
Common mistakes
Check your understanding
Desk assignment

Write a falsifiable question #

Replace “gold looks strong” with a question that a chosen dataset and time horizon can address. A study might ask whether a predeclared curve/flow state changes the distribution of subsequent returns or improves a hedge. The dataset must represent information available at the decision time.

Keep a narrative ledger: observation, interpretation, competing explanation, test and invalidation. A research committee should be able to reject the conclusion without arguing about the analyst’s confidence.

Design the evidence packet #

Include source, observation date, public availability, units, revisions and coverage. Separate ETF tonnage, vault stocks, positioning and market prices rather than summing them into an invented demand statistic. Report which measurements overlap and which are unobserved.

Preserve the original data vintage where licensing and storage terms allow. If historical vintages are unavailable, explain the bias. A polished chart is not an excuse to imply the analyst possessed information earlier than it existed.

Use scenarios before probabilities #

Start with adverse combinations: metal falls while the rupee strengthens; local basis widens against a hedge; lease/funding conditions change; volatility and liquidity move together. Specify price, time, cash requirement and the action if an operational control fails.

Assign probabilities only if you can defend their estimation and uncertainty. A three-row bull/base/bear table with arbitrary percentages is a discussion aid, not a calibrated risk distribution. Expected P&L cannot replace a loss-capacity limit.

Evaluate implementation and the no-position alternative #

State instrument, venue, permission/access assumptions, liquidity, cost, funding and settlement. A relative-price view should not automatically be expressed through an outright leveraged future. Compare the alternative exposures and doing nothing under the same objective.

Define entry eligibility, thesis invalidation and exit timing before paper testing. Do not present a theoretical spread as a trade available to every learner. If a prerequisite is unresolved, label the research incomplete rather than filling the gap with a disclaimer.

Require a decision record and a later review #

The memo should end with accept for paper testing, revise, or reject, with reasons and a scheduled review trigger. Record dissent and what evidence would change the decision. A favourable subsequent price move does not retroactively validate an untestable thesis.

At review, compare process, model prediction, realised path and execution separately. Preserve rejected hypotheses and changed versions. The objective is an auditable research process, not a retrospective collection of correct market stories.

Hypothetical worked example

A fictional scenario memo

Assume a base rupee metal value of 1 million, excluding premiums. A metal return of −8% and USD/INR return of −3% produce 0.92 × 0.97 − 1 = −10.76%, or −107,600. Separate a local-cost stress and derivative hedge outcome rather than adding them into this identity without a convention.

With three scenarios carrying losses of 107,600, 15,000 and a gain of 40,000, no expected-value number can be computed responsibly until probabilities are specified. Even a positive expected value would not make the 107,600 stress affordable.

Common mistakes

Assigning scenario probabilities because the table needs to sum to 100%.
Removing rejected research from the record.
Calling a correct outcome proof of a sound decision.

Check your understanding

An untestable bullish memo is followed by a gold rally. What has that outcome established?

Make this chapter useful

Can you defend the reasoning?

Before moving on, answer the self-check in your own words and write why each alternative fails. Use the worked example to test that explanation. Then apply it in the module research task; keep the assumptions and the evidence that could change your answer.

Desk assignment

What to submit

Submit a two-page committee memo: evidence, mechanism, countercase, scenario losses, cash path, implementation gates and a documented decision.

Use the research ledger for dated evidence, or the module workbooks for calculations. Separate observations, assumptions and interpretation.

Sources and further reading

Further-reading references from the curriculum source list. Contract sizes, expiry calendars, settlement references, margin parameters and tax treatment can change; verify the current operative documents for the actual product. Worked figures are illustrative, not current quotes.

Progress is stored only on this browser.

Back to Gold: market structure and portfolio risk · Learning library · Calculators

General financial education. No securities recommendation, assured outcome or personal advice. Read the risk disclosures.