Intermediate Statistician (China): Integrated quantitative and business foundations

Statistical reasoning: population, uncertainty and defensible reporting

Statistical work begins with a question, a population and a definition of the variable. Decide whether a reported quantity is a count, a rate, a total, an index or a distribution summary. A consistent definition and reference period are essential when comparing organizations or years. Larger files can improve precision under a sound design while still preserving a systematic coverage or measurement error. A result's usefulness depends on how it was obtained, not only on the number of digits displayed.

Sampling and representation

In a simple random sample, the selection mechanism supports specified probability statements about samples. Stratification can improve precision when the grouping and allocation suit the population; a cluster design instead selects groups and needs its own analysis. Random selection and random assignment serve different purposes. A voluntary survey can omit important parts of the population, and a high response count does not automatically remove that selection bias. Identify nonresponse and measurement limitations alongside sampling variability.

Describe before inferring

A mean, median and weighted mean are not interchangeable. State the weights and denominator, and check the distribution for skewness or unusual observations. A sample standard deviation describes spread in observed values, while a standard error concerns the sampling variability of an estimator under its model. A change from one index level to another should be distinguished from a percentage change relative to the starting level. Nominal and real comparisons need an appropriate price measure and consistent base normalization.

Interpret uncertainty

A confidence procedure has a long-run coverage property under its assumptions; an observed frequentist interval does not assign that percentage probability to a fixed parameter. A p-value concerns the compatibility of the data with a specified null model and test statistic, not the probability that the null hypothesis is true. Statistical significance does not by itself measure practical importance, causation or data quality. Check the conditions for a test or interval and communicate important limitations rather than relying on one threshold alone.

Explain patterns and accounts

Regression summarizes a relationship under a fitted model. Confounding, reverse causation, selection and extrapolation can limit a causal interpretation. Economic statistics also require a distinction between stocks at a date and flows during a period, between final output and intermediate transactions, and between a financial position and operating performance. A balanced ledger does not prove that every event was recorded correctly. Preserve definitions, provenance and the reasoning behind adjustments. These foundations support statistical and business reporting without inventing current national classifications, survey powers or legal reporting duties.

Double entry, reconciliation, errors and recording models

Concepts and financial recording

What the domain tests

This domain examines whether you can name the concept behind a treatment, apply it to a year-end adjustment, and distinguish a supplied classroom model from the actual reporting framework. Typical questions ask you to identify the concept being breached or explain why an item is treated as it is.

Core concepts and conventions

Business entity: the owner's private transactions are excluded from the business accounts.
Going concern: accounts assume continued operation, so assets are not stated at break-up values.
Accruals (matching): income and expenses belong to the period they relate to, not the period cash moves.
Consistency: the same method is used each year unless a change is justified and disclosed.
Prudence: exercise caution in estimates under uncertainty without deliberately overstating liabilities or understating assets. The recognition criteria still matter.
Realisation: revenue recognition follows the supplied earning or performance requirements, rather than simply the date cash arrives or an invoice is printed.
Materiality: an item is adjusted if its omission or misstatement could affect a user's decision.
Historical cost: assets are recorded at transaction cost, then depreciated or impaired.

Regulatory framework

Accounting requirements depend on the entity, jurisdiction and specified framework. Begin by identifying the applicable version, entity type and purpose of the statements, then distinguish a mandatory recognition rule from a chosen accounting estimate. International concepts can support comparison without making one country's chart of accounts or tax code binding elsewhere. For a supplied classroom exercise, follow its stated conventions consistently and explain assumptions. A standard's title does not mean every possible asset is measured at historical cost or every event is recognized immediately. Changes in policy, changes in estimates and correction of errors need different analysis under the applicable framework. These chapters teach common financial-recording mechanisms, not a complete current statutory-compliance manual.

Worked example

A trader's year ends on 31 December 2024. December electricity of 180,000 FCFA is invoiced but unpaid. Insurance of 240,000 FCFA was paid on 1 October 2024 for twelve months.

Electricity: Dr Electricity expense 180,000; Cr Accrued expenses 180,000. Matching places the cost in 2024 although cash moves in 2025.
Insurance: only 3/12, or 60,000 FCFA, relates to 2024; 180,000 is prepaid. Dr Prepaid insurance 180,000; Cr Insurance expense 180,000, leaving a 2024 charge of 60,000.

Traps

Prudence is not deliberate understatement; it must be rational. Drawings are not an expense. Consistency cannot defend an error, because a required change or a correction must be considered under the applicable reporting framework. An accrual is not the same as a prepayment: one is a liability, the other an asset.

What the domain tests

This domain tests whether you can record transactions from source documents into the books of prime entry, post them to the ledger, and explain the dual aspect. Exam tasks include journal entries, trade and cash discount calculations, and correcting postings.

Rules and records

Every transaction has two effects: debits increase assets and expenses, credits increase liabilities, income and capital. The reverse reduces them.

The books of prime entry are:

Sales day book: credit sales invoices.
Purchases day book: credit purchase invoices.
Sales returns day book: credit notes issued to customers.
Purchases returns day book: credit notes received from suppliers.
Cash book: cash and bank receipts and payments, often with discount columns.
Petty cash book: small payments, usually under the imprest system.
Journal proper: opening entries, corrections, depreciation, disposals and other non-routine items.

Source documents include invoices, credit notes, debit notes, receipts, petty cash vouchers and bank statements. In the separate-discount-account classroom model used here, trade discount reduces the recorded invoice amount and cash discount is recognized on settlement. Actual revenue, purchase-cost and early-settlement accounting depends on the specified framework.

Worked example

Goods with a list price of 500,000 FCFA are sold on credit at a 10% trade discount. The customer later returns goods with a list price of 50,000 FCFA.

Sale: net invoice value is 450,000. Dr Receivables 450,000; Cr Sales 450,000. The sales day book records 450,000.
Return: trade discount still applies, so the credit note is 45,000. Dr Sales returns 45,000; Cr Receivables 45,000. The customer now owes 405,000. The trade discount never appears in a ledger account.

Traps

Do not post an item to the day book and then post it twice to the ledger: the day book is a list, not a second ledger. A credit note from a supplier goes in purchases returns, not sales returns. A cash-to-bank transfer is recorded on opposite sides of the relevant cash-book columns; a customer/supplier set-off is a different use of the term contra. Identify which transaction is intended. The discount columns in the cash book are memorandum columns and are not part of the double entry until posted.

Errors, reconciliation and financial controls

What the domain tests

You must extract a trial balance, classify errors by whether they affect its agreement, journalise corrections and clear a suspense account. Questions often supply a difference and a list of discovered errors.

Core rules

A trial balance lists debit and credit balances at a date. It proves only that total debits equal total credits, not that the ledger is correct.

Errors that do not affect trial balance agreement:

error of omission, where a transaction is completely left out;
error of commission, where the right class of account but wrong account is used;
error of principle, such as capital expenditure posted to repairs;
error of original entry, where a wrong amount is posted consistently to both accounts;
error of reversal, where the correct accounts are used but the sides are reversed;
compensating errors, which cancel out.

Errors that do affect agreement: one-sided postings, partial omission, different amounts on the two sides, and addition errors in one account.

A suspense account is opened for the difference. Each correction is journalised and the suspense account is cleared when every error is found.

Worked example

The trial balance credit column exceeds the debit column by 36,000 FCFA, so a suspense account with a debit balance of 36,000 is opened. Three errors are found:

The payables entry for a 30,000 credit purchase was correct, but only 3,000 was debited to Purchases. Purchases alone is understated by 27,000: Dr Purchases 27,000; Cr Suspense 27,000.
The customer account was correctly credited for a sales return of 18,000, but the debit to Sales returns was omitted: Dr Sales returns 18,000; Cr Suspense 18,000.
Rent paid of 45,000 was posted to the rent account as 54,000. Rent is overstated by 9,000: Dr Suspense 9,000; Cr Rent 9,000. Suspense account: opening debit 36,000, credits 27,000 and 18,000, debit 9,000. The account clears exactly.

Traps

An error of principle never affects trial balance agreement, so do not look for it as the cause of a difference. Do not correct errors by altering the original source document. A suspense account is temporary, not a real asset or liability. If a difference remains, investigate and document unresolved errors rather than inventing an asset or liability to force agreement. Any real reporting treatment depends on the facts, materiality and applicable requirements.

What the domain tests

This domain tests reconciliation of the cash book bank balance with the bank statement, and the preparation of receivables and payables control accounts. You must distinguish timing differences from errors and know which items enter the adjusted cash book.

Bank reconciliation rules

Items appearing in the bank statement but not the cash book are entered in the adjusted cash book: bank charges, interest, standing orders, direct debits, dishonoured cheques, and dividends or transfers credited by the bank. Items in the cash book but not yet on the statement are timing differences: unpresented cheques and uncredited deposits. Errors by the trader are corrected in the cash book; errors by the bank are notified to the bank.

The reconciliation is: balance per adjusted cash book + unpresented cheques − uncredited deposits = balance per bank statement.

Control accounts

Receivables (sales ledger) control: opening debit balance + credit sales − receipts − discount allowed − sales returns − bad debts + interest charged = closing debit balance.
Payables (purchase ledger) control: opening credit balance + credit purchases − payments − discount received − purchases returns = closing credit balance. Contra entries reduce both control accounts. A credit balance in the receivables ledger is a payable and is shown separately.

Worked example

Cash book bank column shows a debit balance of 500,000 FCFA. The statement shows bank charges of 20,000 and a standing order of 10,000, neither in the cash book. Unpresented cheques total 120,000; deposits not yet credited total 80,000.

Adjusted cash book: 500,000 − 20,000 − 10,000 = 470,000.
Bank statement: 470,000 + 120,000 − 80,000 = 510,000. The 10,000 difference between the original cash book and the statement is explained entirely by the two charges and the two timing items.

Traps

Do not put unpresented cheques in the adjusted cash book; they are timing differences. A dishonoured cheque reduces the cash book and is also reversed in the customer's account. Use a consistent signed-balance or clearly labelled overdraft convention throughout the reconciliation. Financial-statement presentation is a separate question. In control accounts, opening balances must be on the correct side, and contra entries affect both controls.

National accounts, money and fiscal reasoning

National Income, Macroeconomic Objectives and the Circular Flow

Aggregate models connect circular flows, national-account measures, real versus nominal values and macroeconomic objectives. Use the supplied data and accounting conventions rather than treating a country example as current official statistics.

The circular flow

Households supply factors of production and receive wages, rent, interest and profit; firms produce goods and services bought by households.
Withdrawals: saving, taxes, imports. Injections: investment, government spending, exports. Equilibrium occurs where injections equal withdrawals.
The multiplier k = 1 ÷ (MPS + MRT + MPM), where the terms are the marginal propensities to save, tax and import. A more open economy has a larger MPM and therefore a smaller multiplier.

Measuring national income

Three equivalent methods: output (sum of value added), income (factor incomes) and expenditure (C + I + G + X − M).
Exclude transfers as payments for output, the resale value of existing goods and purely financial-asset trades. Separately produced services such as sales commissions can count. National-account production boundaries specify which imputed and own-produced outputs are included; not every unpaid household activity is included.
Real GDP adjusts for price change: real GDP = nominal GDP ÷ price index × 100. The GDP deflator is nominal ÷ real × 100.

Objectives and conflicts

Growth, full employment, price stability under the objectives specified by the relevant institution, balance of payments equilibrium, equitable distribution and environmental sustainability.
Conflicts: the short-run Phillips curve trade-off between inflation and unemployment; growth versus environmental quality; redistribution versus incentives.

Worked example

Nominal GDP rises from €20.0 bn to €21.2 bn, a 6 per cent increase, while the deflator rises from 100 to 102.5. Real GDP in the second year is 21.2 ÷ 102.5 × 100 = €20.68 bn, so real growth is about 3.4 per cent; roughly 2.6 percentage points of the apparent growth was inflation. Separately, with MPS 0.10, MRT 0.15 and MPM 0.25, k = 1 ÷ 0.50 = 2, so a €300 m rise in government investment raises national income by about €600 m before capacity constraints bite.

Traps that cost marks

Treating transfer payments such as pensions as part of GDP; they are not payments for current output.
Confusing GDP (output produced within the territory) with GNP (income earned by residents).
Comparing nominal GDP across years to measure growth.
Applying the full multiplier when the economy is already at full capacity, so the effect appears in prices rather than output.

Money, Banking and Monetary Policy

Money, banking and policy institutions depend on the currency and regulatory regime. Distinguish balance-sheet mechanics and simplified textbook calculations from an empirical claim about a named country’s current monetary system.

Money

Functions: medium of exchange, unit of account, store of value, standard of deferred payment.
Characteristics: general acceptability, divisibility, portability, durability, homogeneity, relative scarcity.
Measures: narrow money (currency in circulation plus overnight deposits) through broader aggregates that add time deposits and money-market instruments.

Credit creation

Credit creation and a textbook multiplier

A commercial-bank loan normally creates a corresponding deposit in the borrower’s account. Lending is constrained by credit risk, capital, funding, liquidity, profitable demand and policy conditions; banks do not simply wait for a fixed pile of customer deposits and relend it mechanically. Reserves serve settlement and regulatory purposes within the relevant system.

The expression 1/r belongs to a deliberately simplified deposit-expansion model with a fixed reserve ratio, full redepositing, no cash leakage, no excess reserves and sufficient willing borrowers. It is a conditional arithmetic model, not a universal prediction of actual loan creation.

Institutions and transmission

A central bank or monetary authority operates under a stated mandate and currency regime. Currency unions, currency boards and independently floating currencies constrain national choices differently. Policy instruments can include administered rates, asset transactions, reserve arrangements and communication. The operational framework may use a corridor or a floor; its actual details must be checked rather than copied from another jurisdiction. Policy changes can affect market funding conditions, bank lending rates, asset prices, expectations and exchange rates. Retail loan rates are not identical to a policy rate, and transmission can be delayed or incomplete.

Worked example

For a hypothetical 1/r deposit-expansion exercise, suppose 5,000 of new reserves enters a closed banking system, r=0.10, every loan is redeposited, no cash or excess-reserve leakage occurs and credit demand is sufficient. The model’s total deposit expansion is 5,000/0.10=50,000, of which 45,000 is additional modeled lending. This is not an observed multiplier for a real banking system. With hypothetical nominal interest 4.25% and inflation 1.8%, i−π gives an approximate real rate 2.45%; the exact gross-factor calculation is 1.0425/1.018−1, approximately 2.41%.

Traps that cost marks

Claiming the ECB sets the rate on mortgages and business loans; it sets policy rates, which feed through.
Assuming banks lend out every euro deposited, ignoring reserves and cash leakages.
Assuming every jurisdiction has an independent national interest-rate or exchange-rate instrument.
Confusing the money multiplier with the Keynesian spending multiplier.

Fiscal Policy, Public Finance and the Budget

Fiscal reasoning distinguishes taxes, expenditure, budget flows, debt stocks, stabilization and distributional choices. All numerical rates and budgets in the exercises are hypothetical givens, not current national tax or fiscal-law claims.

Taxation

Direct taxes fall on income and profit; indirect taxes fall on spending, for example consumption taxes and excise duties under the actual jurisdiction’s definitions and rates on fuel and tobacco.
Progressive, proportional and regressive structures; average versus marginal rate; the Laffer curve shows revenue may fall if rates are pushed too high.
Criteria for a good tax: equity (ability to pay, horizontal and vertical), efficiency, certainty, convenience and economy of collection.

Expenditure and the budget

Current spending on goods and services and public-sector pay; transfers and social benefits; capital investment; interest on the debt.
Budget balance = revenue − expenditure. A deficit adds to the debt stock; the debt-to-GDP ratio also moves with growth, inflation and the interest rate.
Automatic stabilisers (tax receipts and welfare spending that move with the cycle) versus discretionary measures.

Stance and evaluation

Expansionary: higher spending or lower taxes, raising aggregate demand, subject to crowding out and import leakage.
Contractionary: used to control inflation or correct a deficit.
Supply-side measures aim at incentives rather than demand.
Fiscal rules may constrain deficit or debt paths, but their coverage, definitions and escape clauses are jurisdiction-specific. Use only a rule explicitly supplied in the scenario; do not infer legal compliance from a generic budget calculation.

Worked example

A hypothetical budget has revenue 10.5 bn and expenditure 11.2 bn, giving a deficit 0.7 bn. Relative to GDP 30 bn, this is approximately 2.33% of GDP; no actual fiscal-rule compliance is inferred. In a separate closed-economy fixed-price model with MPC 0.8 and spare capacity, the spending multiplier is 1/(1−0.8)=5, so spending 400 m raises modeled income 2 bn. The lump-sum tax multiplier is−0.8/(1−0.8)=−4; a tax cut means ΔT=−400 m, yielding a modeled income increase 1.6 bn. Imports, financing responses, prices, capacity and expectations can alter real outcomes.

Traps that cost marks

Confusing the deficit (a flow over the year) with the debt (a stock at a point in time).
Ignoring automatic stabilisers when judging the discretionary stance.
Applying the full multiplier in a small open economy with high import leakage.
Treating a regressive tax as necessarily unfair in every context; evaluation earns marks.

Economic data, indices and statistical reasoning

Quantitative economic reasoning

Begin by identifying the variable, unit, time period, population and comparison being made. A total, a per-person amount, a percentage share and a rate of growth answer different questions. An arithmetic mean adds values and divides by their number; a weighted mean also specifies the weights and their sum. A median is based on ordered position. An extreme value can influence the mean substantially, so a claim about the typical observation should be checked against the distribution rather than inferred from one average alone.

A price index compares a stated basket or aggregation across periods. In a fixed-base-quantity model, value the same quantities at new and base prices before forming the ratio and multiplying by 100. An index rising from 125 to 130 increases by five index points but by 4 percent relative to its original level. Nominal expenditure divided by an appropriate price index, with its base normalization accounted for, produces a corresponding real measure. The chosen index must match the question; a GDP deflator and a household consumer basket are not interchangeable merely because both measure prices.

State the elasticity convention. An initial-base finite percentage calculation, a midpoint arc elasticity and a point elasticity can give different numerical values. For a midpoint calculation, divide each change by the average of its two endpoint values. This symmetric convention can help relate finite price and quantity changes to revenue comparisons, but it does not prove a constant elasticity across a whole demand curve. A fitted straight line has constant slope while its point elasticity generally varies with price and quantity.

Variation also needs a definition. A population variance and the usual sample variance have different denominators; use the formula supplied. A standard error describes the sampling variability of an estimator under its model, not the spread of every individual observation. If a confidence-interval exercise supplies an estimate, standard error and critical value, calculate estimate plus or minus critical value times standard error and state the assumptions. An observed frequentist interval does not assign that confidence percentage to the fixed parameter itself.

Correlation and regression summarize patterns. A fitted positive association is not proof that intervening on one variable causes a change in another. Confounding, reverse causation, selection and measurement can matter. Predicting within the observed range differs from extrapolating far beyond it. A voluntary online survey may exclude important groups even if it receives many responses; sample size does not automatically cure selection bias.

Finally, use exact ratios when precision matters. Per-person output growth is the growth factor of output divided by the growth factor of population, minus one. Subtracting growth rates is an approximation. Carry units through every step, avoid unnecessary intermediate rounding and distinguish the arithmetic result from the causal or policy conclusion the evidence can actually support.