Saturday, 10 January 2015

#GE2015 Forecast by ElectionsEtc

Forecast b 150109
Happy New Year! Our first forecast of 2015 finds the polls little changed since our last of 2014. The Conservatives have recovered a point, meaning they now trail Labour by just one point in our polling average, 32% to 33%. The Lib Dems have held steady on 9% and UKIP are down one to 14%.
That means a slight improvement for the Tories in our model’s forecast: their chances of winning the most seats are up from 45% three weeks ago to 48% now; their chances of a majority are up from 16% to 19%.
Our central forecast is for the Tories to win the most votes – with 34% to Labour’s 31.2% –but come out with slightly fewer seats than Labour: 294 to 297 (with the Lib Dems on 29).
However, we have yet to finalise a new method that properly takes into account the SNP’s rise and Labour’s fall in Scotland – we’ll be introducing it in the next week or two. That means that, for the moment, our forecasts come with a big health warning: they are likely to overestimate Labour.
If you want to do a rough-and-ready adjustment, the Labour-to-SNP swing in Scotland probably means around 35 seats off Labour’s total and 5 off the Lib Dems’. Clearly, therefore, the chances of Labour being the largest party are lower than the 52% our current model suggests – in reality, the Scottish situation means the Conservatives are clearly the favourites. Similarly, the likelihood of a Labour majority if probably somewhat lower than the 21% in this week’s forecast, and that of a Hung Parliament is probably higher than 60%.
Once we launch our new method, we’ll continue to run our old method alongside it as a kind of benchmark, showing what things would more typically looked like with GB-wide three party dynamics only.

Date of forecast: 9 January 2015
Days till the election: 118
Inputted current average poll shares
Con: 32%
Lab: 33%
LD: 9%
Others (inc. UKIP): 26%
– UKIP: 14%
Forecast Election Day Shares (with 95% Prediction Intervals)
Con: 34.0% (±6.4, i.e. 28% – 40%)
Lab: 31.2% (±4.8, i.e. 26% – 36%)
LD: 11.4% (±6.0, i.e. 5% – 17%)
Implied point estimate shares for:
– Others (inc. UKIP): 23.4%
– UKIP: 12.6%
Forecast Election Day Seats (with approximate 95% Prediction Intervals)
Con: 294 (225 – 371)
Lab: 297 (223 – 361)
LD: 29 (24 – 35)
(Prediction intervals assume LD & others shares at central forecast, Con & Lab shares vary as per prediction intervals)
Central forecast: Lab largest party, but short of a majority by 29
Approximate probabilities of key outcomes
Con largest: 48%
… with a majority: 19%
Lab largest: 52%
… with a majority: 21%
Hung Parliament: 60%
… with Con largest: 30%
… with Lab largest: 31%
(probabilities may not sum due to rounding)

No comments:

Post a comment

Comment is open to all feel free to link to this blog.