We have been reporting on the performance of our share price model for Loews Corporation (NYSE:L) since 2010. In March 2012, we presented an updated model including data available for February 2012, and predicted a slight price growth in Q2. Actually, the (monthly closing adjusted for splits and dividends) price grew from $39.02 in February to $41.0 in April, with a following slight decrease to 40.85 in June. Here, we update the model with the closing price through October 2012 and the (not seasonally adjusted) consumer price indices for September 2012. We have found that the share price may grow by $6 in the December 2012-January 2012 timeframe, and definitely might be considered as a potentially good investment idea. The standard deviation of our model since June 2003 is only $2.43, and the increment of $5 is definitely above the uncertainty bounds. At the same time, there were short periods of large deviation between the predicted and observed curve, e.g. in 2010.
The model is based on the decomposition into a weighted sum of two consumer price indices (selected from a larger set of CPIs), linear trend and constant; all coefficients and time lags to be estimated by a LSQ procedure. A month ago, we presented a quarterly report and confirmed the stability of the original model obtained in September 2009 for the period through October 2008. Here, we test the previous model and make a regular update using new data. All in all, the original model is valid since October 2008, and does not show any clear sign of changes in the future. This is a reliable model, valid during the past 60 months!
A preliminary model for Loews Corp. was obtained in September 2009 and covered the period from October 2008. This old model included the index of food without beverages (FB) which led by six months, and the index of transportation service (TS) with a four month lead:
L(t) = -2.52FB(t-6) - 1.38TS(t-4) +27.93(t-1990) + 377.24, stdev=$2.04, September 2009
where L(t) is the share price in U.S. dollars, t is calendar time.
Since November 2010, the defining indices were the same: the index of food and beverages (F) and the TS index. Figure 1 depicts the evolution of the indices that provide the best fit model, i.e. the lowermost RMS residual error, between July 2003 and October 2012. The food and beverages index leads by five months, and the TS index by four months. The model does not show any tangible change with time -- only coefficients have been slightly fluctuating:
L(t) = -2.04F(t-5) - 2.08TS(t-4) +28.09(t-1990) + 441.81, Nov. 2010
L(t) = -2.03F(t-5) - 2.12TS(t-4) +28.23(t-1990) + 448.98, March 2011
L(t) = -2.01F(t-5) - 2.09TS(t-4) +27.96(t-1990) +440.65, Sept. 2011
L(t) = -2.03F(t-5) - 2.02TS(t-4) +27.65(t-1990) +431.99, Dec. 2011
L(t) = -2.01F(t-5) - 2.01TS(t-4) +27.49(t-1990) +428.70, stdev=$2.41, Feb. 2012
L(t) = -2.00F(t-5) - 1.96TS(t-4) +27.18(t-2000) +693.99, stdev=$2.43, Sept. 2012
The current model is depicted in Figure 2, together with high and low monthly prices as a proxy to the uncertainty bound of the share price. The predicted curve leads the observed one by four months. The solid red line presents the contemporary prediction, i.e. one sees four months ahead. Major falls and rises are well forecasted four months in advance. It is worth noting that the model obtained in March 2011 accurately predicted the small fall observed in the second and third quarters of 2011. The model residual error is of $2.43 for the period between July 2003 and October 2011, as shown in Figure 3.
Figure 1. Evolution of the price indices F and TS.
Figure 2. Observed and predicted share prices.
Figure 3. The model residual error; stdev=$2.43