Private Wealth Management interview preparation

Client discovery, goals-based planning, asset allocation, tax and estate structuring, products and the commercial reality of building a book, with substantial Indian content on PMS, AIFs, SEBI's adviser rules and family structures. Every question is either traced to a named firm from a public candidate report, or tagged at desk level when we could not trace it.

Jump to the question bank
Question bank

100 questions, mapped to the firms that asked them

Questions
100
Traced to a firm
22
Firms
13
Updated
September 2026
Asked at
All firmsAllianceBernstein4Goldman Sachs4Northern Trust3J.P. Morgan2Morgan Stanley2Scotiabank2Ares Management1BNY Mellon1Carlyle Group1Invesco1Neuberger Berman1Schroders1UBS1
Topic
All topicsClient discovery5Risk profiling4Asset allocation and rebalancing7Investment policy statement3Tax and asset location6Concentrated positions3Estate, succession and philanthropy6Fiduciary and trusts3Alternatives and liquidity4Products and platforms7Fees and conflicts4Bank economics and risk2Behavioural finance3Family governance3Onboarding and compliance3Business development6Fit and career15Markets and economy9Case and estimation7
Level
AnyCoreIntermediateHard
Type
AnyTechnicalCaseMarket viewFitBrainteaser
Showing 1–1 of 1 · filtered from 100Clear filters
  1. 012Walk me through mean-variance optimisation, and tell me why you would not hand the output to a client.Asset allocation and rebalancingHardsuperdayFamily officesWealth management

    Say this

    You feed in expected returns, volatilities and correlations, and it gives you the mix with the highest expected return for a given volatility. The problem is that it is an error-maximiser: tiny changes in the expected return inputs produce wildly different and usually absurd portfolios.

    Then walk it

    1. The mechanics: for each level of risk, the optimiser finds the weights that maximise expected return, and the set of those points is the efficient frontier. You then pick the point that matches the client's risk budget.
    2. The first failure is input sensitivity. Expected returns are estimated with huge error, and the optimiser loads up on whichever asset you happened to be most optimistic about. Michaud called it error maximisation and the name is fair.
    3. The second failure is corner solutions. Unconstrained, it will hand you 40 percent in emerging market small caps and zero in domestic large caps, which no client will hold and no committee will approve.
    4. The third is that correlations are unstable and rise in crises, which is exactly when diversification is supposed to pay. The matrix you optimised on is a fair-weather matrix.
    5. What I would actually do: use it as a diagnostic, not a decision. Constrain the weights to sensible ranges, use reverse optimisation or a Black-Litterman approach so the starting point is the market portfolio rather than my own return forecasts, and resample to see how stable the answer is.
    6. And the fourth failure, the one that matters most for a private client: variance is not the risk the client cares about. It ignores taxes, illiquidity, drawdown path and the fact that he may sell at the bottom. A portfolio that is 30 basis points off the frontier but that he will hold beats the optimal one he abandons.

    Where candidates lose it

    Describing the frontier competently and stopping. The question has 'why would you not hand it to a client' in it. Name error maximisation and the fact that variance is not the client's risk measure, or you have answered half the question.

    Expect next

    • What is Black-Litterman doing differently?
    • How do you handle illiquid assets in an optimiser?
    • What constraints would you impose and why?

Firm tags come from public, anonymous candidate reports on Wall Street Oasis: strong signal, not sworn testimony. Firms are named as the places a question was reported, not as partners of Fin Maverick. Answers are written for this page to show how to think out loud; they are not scripts to recite.