Sensitivity Analysis to Assumptions

Sensitivity analysis is a systematic technique that measures how the output of a valuation model or return estimate changes when one or more input assumptions are varied. It isolates the effect of each assumption, allowing the analyst to see which inputs drive the result the most.

In the context of the NISM research analyst exam, sensitivity analysis is used to test the stability of projected returns, Net Present Value (NPV), or earnings per share (EPS) against changes in discount rates, growth rates, beta, expense ratios, and other key drivers.

The exam frequently presents a base‑case scenario and asks you to calculate the impact of a ±10% change in a specific assumption. Recognising the logical flow of the analysis saves time and prevents arithmetic errors.

• It highlights the most critical assumptions that require careful scrutiny.
• It supports risk‑adjusted recommendations by showing best‑case and worst‑case outcomes.

Most valuation and return models rely on a handful of quantitative inputs. The most common ones are:

Discount rate – the required rate of return used to discount future cash flows. In Indian equity research, the discount rate often reflects the Capital Asset Pricing Model (CAPM) inputs.

Cash‑flow growth rate – the expected annual increase in earnings or dividends. Small changes here can swing the projected return dramatically.

Beta (β) – measures the security's systematic risk relative to the market index. SEBI’s guidelines require analysts to disclose the source of the beta estimate.

Expense ratio – for mutual funds, the annual cost expressed as a percentage of assets under management. A higher expense ratio reduces net returns.

Other assumptions include tax rates, terminal growth rates, and currency conversion factors. Knowing which of these are most volatile helps you prioritize the sensitivity tests.

Step 1 – Establish a base‑case model using the most realistic assumptions. Record the base output (e.g., projected return or NPV).

Step 2 – Select the assumption(s) you will test. Typical practice is to vary each input by a fixed percentage, such as ±10% or ±20%, depending on its historical volatility.

Step 3 – Re‑run the model for each variation while keeping all other inputs constant. Capture the new output values.

Step 4 – Calculate the change in output relative to the base case. Present the results in a table or chart that shows the direction and magnitude of the impact.

Step 5 – Analyse the pattern. If a small change in an assumption leads to a large swing in the output, that assumption is deemed highly sensitive and warrants careful justification in your research report.

When the chart or table shows a steep slope, the underlying assumption is a key driver of risk. For exam questions, a steep slope often signals that the analyst must disclose the assumption and possibly recommend a conservative estimate.

If the output changes only marginally across a wide range of the assumption, the model is robust. In such cases, the analyst can place greater confidence in the base‑case recommendation.

SEBI’s research guidelines require a clear statement of the sensitivity range and its impact on the final recommendation. Forgetting this disclosure can lead to a loss of marks in the essay‑type questions.

Sensitivity analysis complements other risk tools such as Value‑at‑Risk (VaR) and stress testing. While VaR quantifies potential loss under normal market conditions, sensitivity analysis shows how a specific assumption shift (e.g., a sudden rise in interest rates) would affect the projected return.

In practice, analysts often run a sensitivity matrix alongside VaR to provide a holistic view of both market‑wide and assumption‑specific risks. The NISM exam may ask you to choose the appropriate tool for a given situation.

Remember that stress testing is an extreme form of sensitivity analysis where the assumption change is large (e.g., a 50% drop in earnings). The underlying methodology, however, remains the same.

When performing sensitivity analysis for Indian equities, use realistic ranges based on historical volatility of the Indian market. SEBI’s research guidelines suggest referencing the past five‑year standard deviation of the index when setting discount‑rate ranges.

For mutual funds, the expense ratio range should reflect the category‑specific norms published by AMFI. Using sector‑specific growth rates (e.g., IT sector 12‑15% CAGR) improves credibility.

Document every assumption change and its source (e.g., RBI policy change, quarterly earnings revision). This documentation is a mandatory part of the research report under SEBI (Research) Regulations, 2011.

1. Build a solid base‑case model with the best available data.
2. Choose key assumptions and define a realistic sensitivity range.
3. Re‑run the model for each variation, keeping other inputs constant.
4. Record the adjusted outputs and calculate percentage changes.
5. Present the results in a clear table or chart and highlight the most sensitive drivers.

Following this checklist ensures you cover all exam requirements and avoid common calculation errors.

Finally, always link the sensitivity findings back to the investment recommendation, stating whether the recommendation remains unchanged, becomes more cautious, or needs a disclaimer.