An article in the Evening Standard understated the amount of hard data in GDP estimates and quoted a City commentator questioning the value of the preliminary estimate. Joe Grice wrote defending the ONS figures.
You say that the Office for National Statistics' second and third estimates of GDP contain only 60 per cent and 80 per cent respectively of the final data ('Recovery may be faster than expected but it will be a long slog', 23 February). This is not correct: the true figures are 77 per cent for the second estimate and 92 per cent for the third estimate. Mr Ostwald, whom you quote, is entitled to his opinion on the value of our preliminary estimate, but it is no 'guesstimate'. Our published research shows that the average revision over the last five years is less than 0.2 per cent and there is no consistent bias up or down.
Office for National Statistics
Newport NP10 8XG