G R INFRAPROJECTS LIMITED (GRINFRA) Price Target & Share Price Forecast

How the GRINFRA price target & forecast are calculated

We ran 10,000 simulated price paths for G R INFRAPROJECTS LIMITED (GRINFRA) using Geometric Brownian Motion (GBM) - the same probability framework used in institutional risk-management systems. A GBM needs exactly two inputs: how much the price moves, and which way it drifts. We treat those two very differently, and the reason is worth a minute of your time, because it is where most published "price targets" go wrong.

Volatility: measured, from the last two years

GRINFRA's annualised volatility is 30.5%. Volatility is easy to measure and it persists: a couple of hundred trading days pin it down to within a few percent of itself, and this year's level is a decent guide to next year's. So we simply measure it, from the recent window, and use it as-is. That number sets the width of every band on this page.

Drift: estimated, then shrunk toward a long-run anchor

Drift is the opposite problem. Over the 5.0 years of history we hold, GRINFRA compounded at -14.0%/year. But the statistical uncertainty on any drift estimated from 5.0 years at 30.5% volatility is about plus or minus 14.5%/year. That is not a rounding error, it is wider than the answer. Compounding a number that noisy over sixty months is how a large, profitable company ends up with a published five-year median below today's price for no reason other than which two dates the window happened to start and end on.

So we do three things instead of extrapolating it:

• Use the middle of the record, not the endpoints. The sample drift here is -5.9%/year, taken as the median of every one of 497 overlapping three-year stretches in the series. One bad quarter at the end cannot set the five-year story.
• Shrink it toward a long-run anchor. The anchor is a 10.9%/year log return, which is a ~6.5% government-bond yield plus a ~5% equity risk premium, minus GRINFRA's own measured dividend yield of 0.3% (this page forecasts the price, and dividends are paid out of it) and minus its volatility drag of 4.7% (a median is not a mean; volatility pushes them apart). That gives an anchor of 5.9%/year. The weight on it is set by how noisy the sample is - here 89% anchor, 11% this company's own record.
• Cap the result at a defensible band of -15% to +20% a year, so a short, wild history cannot publish an absurd compounding rate. For this symbol the cap is not binding.

The drift actually simulated is 4.7%/year. Every one of the 10,000 trials projects a day-by-day path from today's ₹871 using that drift and that volatility, and the P10/P50/P90 bands are simply where those paths land.

What this cannot tell you

A GBM assumes volatility stays put and that returns are normally distributed. Real markets do neither: they cluster, they gap, and they crash further than a bell curve allows. Read P10 as a soft floor, not a worst case - a genuine bear market can go through it. The model also knows nothing about G R INFRAPROJECTS LIMITED's earnings, balance sheet, competition or management; it reads price and nothing else. It is a way of sizing uncertainty honestly, not a view on the business.

Why this differs from an analyst price target: analyst targets are point estimates from subjective valuation models. This is a probability distribution from measured market data plus a stated, deliberately conservative long-run anchor. It tells you the range and the likelihood, not a number to bet on.