For your information
Joint meeting with Business and Industrial Section
------------------------------------------------------------
Date: Wednesday 21 April
Venue: ONS, Newport [directions from main gate]
Programme: (abstracts at end)
2.30 Jake Ansell (Edinburgh) "Can we manage risk?"
3.30 Tea
4.00 Mick Silver (Cardiff) "Estimating hedonic regressions for
durable goods"
ONS, Newport is close to J28 on the M4 - further directions and/or
information from
Alan Watkins: [log in to unmask] 01792 295853
Paul Smith: [log in to unmask]
Abstracts
------------
Jake Ansell: Can we manage risk?
----------------------------------------
Every decision has a possible downside, risk. The issue is whether we
can manage risk. Clearly, as statisticians, we hope to be able to
measure the risk. In many areas, people believe they are using
appropriate measures to manage risk. By looking at a range of
examples, it will be shown that management is not easy and that
frequently used measurements provide risks themselves. The need is
for more statisticians to be involved with the management of risk.
Mick Silver: Estimating hedonic regressions for durable goods
---------------------------------------------------------------------
Hedonic regressions are simply the regression of the price of a
model of a good (for example a PC, TV or dishwasher) on each of its
quality characteristics. The coefficients are argued to provide
estimates of the marginal value consumers attribute to each
characteristic, which may include the make of the model and its year
of introduction. They have been used to estimate quality-adjusted
price changes by including in the specification dummy variables for
the time period the model is sold in. The results of such studies
have been used as indicators of bias due to quality changes in
consumer price indices. However, such formulations can be misleading
and inappropriate for the estimation of such bias. With the help of
bar-code scanner data it will be shown how superlative hedonic
quality-adjusted price changes improve on these estimates. there will
also be some discussion of how all of this relates to
quality-adjustment methods for consumer (and producer) price indices
as well as some economic issues and results.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|