Quarterly report on Boston Scientific

We posted on Boston Scientific (BSX) in January, April, and October 2011 and presented a share price model for Boston Scientific as based on our stock pricing concept.  Both models were similar and included the consumer price index of housing (H) and the index of durable goods (DUR). (Figure 1 depicts the overall evolution of the involved indices.) In the currently re-estimated model (i.e. through December 2011), the former defining CPI component led the share price by 6 month and the latter one by 4 months (slightly increased lags but contemporary CPIs are used in the December model).

  

Thus, here we update the original model using data through December 2011. The updated model has the same defining components and time lags with slightly different coefficients. Therefore, the original model provided a reliable prediction through the past year and further in the past. Currently, the best fit model predicts the share to fall below zero in the near future what is equivalent to bankruptcy. A similar prediction (negative share price) was obtained for Lehman Brothers and other financial institutions before they failed. 

The best-fit 2-C models (March, September, and December 2011) for BSX(t) are as follows: 

BSX(t) = -1.41H(t-5) – 2.85DUR(t-3) – 0.09(t-1990) + 630.92, March 2011

BSX(t) = -1.46H(t-5) – 2.71DUR(t-3)  + 0.36(t-1990) + 615.69, September 2011  

BSX(t) = -1.44H(t-6) – 2.21DUR(t-4)  + 0.84(t-1990) + 546.90, December 2011

where BSX(t) is the (monthly closing adjusted for splits and dividends) share price in US dollars,  t is calendar time.  Both coefficients are negative, and thus the increasing consumer prices result in decreasing share price. The slope of time trend is almost negligible, i.e. $0.84 per year.  

All models predicted the price at a three-to-four month horizon with standard deviation of $2.24 between July 2003 and December 2011 ($1.83 in March). The currently observed growth in the defining consumer price indices should drive the share price down. In the first quarter of 2012, the price may still drop below zero.  However, the price will gradually recover above the zero line. The model residual in Figure 3 should return to the zero line somehow. 

Figure 1. The evolution of H and DUR. 

Figure 2. Observed and predicted BSX share prices. March, September, and December 2011 models. The price is negative by the end of 2011.

Figure 3. The model residual, i.e. the difference between the observed and predicted BSX share prices.

Tim Duy on Japan or why macroeconomics is wrong

Tim Duy on Economists View posted (http://economistsview.typepad.com/economistsview/2012/01/fed-watch-japan-revisited.html) on the current recovery of Japan and also mentioned usuall macroeconomic rubbish on "lost decades". This is one of good examples showing that the mainstream macroeconomic theories are  worthless and confusing. These people do not actually understand what drive a developed economy like the Japanese one.

We have already decribed the evolution of  a developed economy as expressed by real GDP per capita, G. There are two component in play - inertial growth, A/G,  and the change in a specific age population, dNs/Ns ( following paragraph is borrowd from our book "mechanomics. Economics as Classical Mechanics)

dG(t)/G(t)=A/G(t)+0.5dN_s(t)/N_s(t)dt   (1.8)

where A is an empirically determined coefficient, N_s(t) is the number of people of the defining age. For Japan, the defining age of eighteen years has been found.  Relationship (1.8) implies that the growth rate of GDP depends explicitly and entirely on the attained level of real GDP per capita and the population change. If to gather relevant terms on both sides of the equation, this relationship can be simplified in the following form: 

d[G(t)-(At+C)]/G(t)= 0.5dN_s(t)/N_s(t)  (1.9)

where C is the constant of integration, i.e. the initial condition of the initial value problem. 

From (1.9) one can derive either the evolution of G or Ns depending on the purpose. Since the number of 18-year-olds can be estimated by integrating (actually by summation of discrete estimates) the left hand side of (1.9) with the measured annual values of G and also enumerated by population surveys one can  compare results visually and statisticaly. Figure 1.23 (also from the book) demonstartes that the evolution of G follows up the evolution of Ns. Since G can not affect Ns the causality directio is opposite - the change in Ns drives G.   From 1.23, one can  understand that so called "lost decades" actually manifest the fall in the number of 18-year-olds since 1992. Accordingly, the years before 1991 are charaterized by increasing Ns and thus are called "economic  miracle".  Finally, one can extrapolate the younger age cohorts into the future and estimate the future evolution of G. Figure 1.23 shows that the years of low economic growth are left behind and the 2010s will be charaterized inertial growth only, A/G, since Ns will not be changing. This is not fast economic growth, A/G~ 1.5% per year, but is definitely better than the permanent depression of the 1990s and 2000s.

Figure 1.23. Enumerated and predicted number of 18-year-olds.

On the prostest of Harvard economics students

Actually, it happened two months ago and I've just learnt it from Robert Johnson who posted on the failure of economic and financial theories in 2008 ( http://business.time.com/2012/01/19/economists-a-profession-at-sea/?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+timeblogs%2Fcurious_capitalist+%28TIME%3A+Business%29) . Obviously, there is a trade-off between  theories and observations. Soft sciences, like history or philosophy, do not pretend to  describe the world in quantitative terms. Hard sciences are fully quantitative and do not consider any deviation from the scientific method of proof, i.e. statistics and mathematics. Economics is somehow in-between the soft and hard sciences. Economists base their theories on assumptions from the blue sky but want the outcome to be considered as in the hard sciences. There is a clear and irresolvable conflict which surfaces over and over during large economic perturbations - one  pretends to adequately describe by math and statistics what s/he pretends can not be adequately describe by math and statistics. Macro economists never want and always deny their theories to be tested econometrically, i.e. in scientific way.  Hence, the current situation (i.e. the painful failure to predict large changes) with economics will be repeated over and over because it is embedded in the internal structure of the economic theories. 

Students have always seen and complained on the intrinsic  weakness on the macroeconomic postulates. Essentially, these postulates do not give any appropriate understanding of the real world and thus confuse. (In the end of Ec10 students are usually informed that nothing from they have just learnt actually works.)  The protesting Harvard students just express this simple idea which has always been with them before but which they could not express openly in quiet times due to the concealed threat from their profs.

Our economic concept is based on observations and actually predicts quantitatively the evolution of developed economics.  When these protests gather more supporters demanding scientifics knowledge our concept (fully following the method adopted in the hard sciences) will be more easily accepted as the only alternative to the "soft" economics.

Quarterly report on Aflac share price

Here we routinely revisit the stock price model for Aflac Incorporated (AFL) which we obtained in October 2011 using our concept of share pricing. Accordingly, our goal is to test the original model and to update time lags and coefficients.

Last time we predicted the AFL price using the CPI estimates published by the BLS for September 2011. It was a preliminary model.  The share price was defined by the consumer price index of household furnishing and operations (HFO) and that transportation services (TS). In September, the defining time lags are as follows: the HFO index led the share price by 0 months and the TS by 5 months. Adding new data for the period between September and December 2011, we have re-estimated the model and found some changes in the time lags: one and six months, respectively, as well as in the estimated coefficients. The relevant best-fit 2-C models for AFL(t) are as follows:  

AFL(t) =  -5.02HFO(t-2) – 2.87TS(t-6)  + 20.42(t-1990) + 997.71,  March 2011

AFL(t) =  -4.63HFO(t-0) – 2.90TS(t-5)  + 20.41(t-1990) + 953.49, September 2011

AFL(t) =  -4.63HFO(t-1) – 2.87TS(t-6)  + 20.23(t-1990) + 948.72, December 2011  

where AFL(t) is the AFL share price in U.S. dollars,  t is calendar time. The changes in time lags are shown in red.  

In July 2011, we reported that the original model gave a correct prediction of the fall in Q2 2011. In September 2011,  we showed that the contemporary fall in the price had to stop and expected a positive correction in Q4 2011. Figure 1 confirms our prediction and depicts the high and low monthly prices for an AFL share together with the predicted and measured monthly closing prices (adjusted for dividends and splits). The model residual error is depicted in Figure 2. The price will likely not change much in 2012Q1.     

Figure 1. Observed and predicted AFL share prices.

Figure 2. The model residual error.

A share price for Advanced Micro Devices

Here we address the stock price model for Advanced Micro Devices (AMD) first time. The model has been obtained using our concept of share pricing. Accordingly, our goal is to test the underlying concept and to estimate time lags and coefficients.

We have borrowed the time series of monthly closing prices of AMD from Yahoo.com and the relevant (seasonally not adjusted) CPI estimates through December 2011 are published by the BLS for March 2011.  The evolution of AMD share price is defined by the consumer price index of rent of primary residence (RPR) and that hospital and related services (HOSP). The defining time lags are as follows: the RPR index leads the share price by 1 month and the HOSP by 4 months. The relevant best-fit 2-C models for AMD(t) is as follows:  

AMD(t) =  -2.21RPR(t-1) – 0.82HOSP(t-5)  + 37.87(t-1990) + 267.01,  December 2011

where AMD(t) is the AMD share price in U.S. dollars,  t is calendar time.  Figure 1 depicts the high and low monthly prices for an AMD share together with the predicted and measured monthly closing prices (adjusted for dividends and splits). As a rule, the predicted prices are well within the bounds of the share price uncertainty.  The model residual error is shown in Figure 2. We will be reporting on AMD. 

Figure 1. Observed and predicted AMD share prices. 

Figure 2. The model residual error.

Alcoa Inc. share price through March 2012

Three and nine months ago we reported on a share price model for Alcoa (AA), which is company from Materials subcategory of the S&P 500 list specialized in aluminum. Here we test and update the model using new data, including the monthly closing price in December 2011 and the estimated CPI components for December. The principal result is that the model has the same defining CPI components and time lags with slightly shifted coefficients. Therefore, the model is a reliable tool to predict the evolution of Alcoa at a three month horizon.

According to our general approach to share price modeling we decompose the observed time history of the monthly closing AA stock price (adjusted for splits and dividends) into a weighted sum of two CPI components, time trend and free term.  Two defining CPI components are selected to minimize the model (RMS) error and may lead or lag behind the share.  

The original and current AA model is defined by the (not seasonally adjusted) index of food away from home (SEFV) and the price index of rent of primary residence (RPR), as reported by the US BLS. The former CPI component leads the share price by 3 months and the latter is 5 months ahead of the share price. Figure 1 depicts the overall evolution of both involved indices through August 2011. It seems these indices have been evolving in sync since 2002 with the only step-like change in the SEFV index in 2008.  We present three empirical models as estimated in April, October, and December 2011:  

AA(t) =  -6.71SEFV(t-2) + 3.34RPR(t-4)  + 19.23(t-1990) + 298.87, Aril 2011

AA(t) =  -6.61SEFV(t-2) + 3.22RPR(t-4)  + 19.51(t-1990) + 300.89, October 2011

AA(t) =  -6.47SEFV(t-3) + 3.08RPR(t-5)  + 19.58(t-1990) + 302.45, December 2011

where AA(t)  is a share price in US dollars, t is calendar time. Figure 2 illustrates the observed and predicted models for December 2011. The residual error is $3.05 ($3.04 in October and $3.12 in April) for the period between July 2003 and December 2011. Figure 2 also shows monthly high and low prices as the uncertainty in the monthly closing price as the best share price estimate. Since the closing price has to characterise the whole month by one value the high and low prices might serve as statistical bounds.

One can expect the share price to rise through March 2012.  

 

Figure 1. Evolution of the price of SEVF and RPR.

Figure 2. Observed and predicted AA share prices.  

It is better to see one time than to hear 100 times

I have mentioned that the graph shown by Lane Kenworthy

has a significant problem with data. Here is a graph from my paper from 2005 on average (and median) personal income in the USA.

Fig. 3. Ratio of the total number of people and the number of people with income in various age groups. The youngest group is characterized by a participation factor of 0.75. During the late 1970s, participation factor in other age groups increased from 0.82-0.85 to 0.92 -0.99.

One can see that personal income definition was dramatically revised around 1977 and the portion of people with income jumped by 10% to 15%. The values above 1.0 also manifest the absolutely stupid difference between people counted by the BLS and the Census Bureau. There are to United States according to their statistics. The BLS does not take into account the closure error after decennial censuses and thus uses wrong populations, as I mentioned many times in this blog.

All in all, it is very suspicious that the deviation between GDP per capita and median income started the very same time.