Effects of measurement errors in predictor selection of linear regression model

Measurement errors may affect the predictor selection of the linear regression model. These effects are studied using a measurement framework, where the variances of the measurement errors can be estimated without setting too restrictive assumptions about the measurement model. In this approach, the problem of measurement is solved in a reduced true score space, where the latent true score is multidimensional, but its dimension is smaller than the number of the measurable variables. Various measurement scales are then created to be used as predictors in the regression model. The stability of the predictor selection as well as the estimated predicted validity and the reliability of the prediction scales is examined by Monte Carlo simulations. Varying the magnitude of the measurement error variance four sets of predictors are compared: all variables, a stepwise selection, factor sums, and factor scores. The results indicate that the factor scores offer a stable method for predictor selection, whereas the other alternatives tend to give biased results leading more or less to capitalizing on chance.     

Computing Equilibria in Discounted 2 × 2 Supergames

This article examines the subgame perfect pure strategy equilibrium paths and payoff sets of discounted supergames with perfect monitoring. The main contribution is to provide methods for computing and tools for analyzing the equilibrium paths and payoffs in repeated games. We introduce the concept of a first-action feasible path, which simplifies the computation of equilibria. These paths can be composed into a directed multigraph, which is a useful representation for the equilibrium paths. We examine how the payoffs, discount factors and the properties of the multigraph affect the possible payoffs, their Hausdorff dimension, and the complexity of the equilibrium paths. The computational methods are applied to the 12 symmetric strictly ordinal 2 × 2 games. We find that these games can be classified into three groups based on the complexity of the equilibrium paths.     

A COMPARISON OF HPLC WITH FLUORESCENCE DETECTION AND FIXED WAVELENGTH FLUORESCENCE METHODS FOR THE DETERMINATION OF POLYCYCLIC AROMATIC HYDROCARBON METABOLITES IN FISH BILE

Fish exposed to polycyclic aromatic hydrocarbons (PAHs) accumulate only a small amount of PAH in their tissues. Hydrophobic PAHs are readily transformed into hydrophilic metabolites and excreted in bile. Bile samples of five fish species were collected from the Baltic Sea and analyzed using two methods: high-performance liquid chromatography (HPLC) with fluorescence detection and a fixed wavelength fluorescence method. With the HPLC method PAH compounds were quantitated after hydrolysis, and with the fluorescence method without enzyme-assisted hydrolysis. In HPLC analysis the major metabolite in all fish species, 1-hydroxypyrene, was the only metabolite detected in perch and salmon samples. 1-Hydroxyphenanthrene was detected in a few flounder and in most eelpouts. The correlation between the two methods for pyrene-type metabolites was good (r² = 0.773).     

Modelling opinion formation driven communities in social networks

In a previous paper we proposed a model to study the dynamics of opinion formation in human societies by a co-evolution process involving two distinct time scales of fast transaction and slower network evolution dynamics. In the transaction dynamics we take into account short range interactions as discussions between individuals and long range interactions to describe the attitude to the overall mood of society. The latter is handled by a uniformly distributed parameter α, assigned randomly to each individual, as quenched personal bias. The network evolution dynamics is realised by rewiring the societal network due to state variable changes as a result of transaction dynamics. The main consequence of this complex dynamics is that communities emerge in the social network for a range of values in the ratio between time scales. In this paper we focus our attention on the attitude parameter α and its influence on the conformation of opinion and the size of the resulting communities. We present numerical studies and extract interesting features of the model that can be interpreted in terms of social behaviour.     

Acrylamide levels in Finnish foodstuffs analysed with liquid chromatography tandem mass spectrometry

Sample clean-up and HPLC with tandem mass spectrometric detection (LC-MS/MS) was validated for the routine analysis of acrylamide in various foodstuffs. The method used proved to be reliable and the detection limit for routine monitoring was sensitive enough for foods and drinks (38 microg/kg for foods and 5 microg/L for drinks). The RSDs for repeatability and day-to-day variation were below 15% in all food matrices. Two hundred and one samples which included more than 30 different types of food and foods manufactured and prepared in various ways were analysed. The main types of food analysed were potato and cereal-based foods, processed foods (pizza, minced beef meat, meat balls, chicken nuggets, potato-ham casserole and fried bacon) and coffee. Acrylamide was detected at levels, ranging from nondetectable to 1480 microg/kg level in solid food, with crisp bread exhibiting the highest levels. In drinks, the highest value (29 microg/L) was found in regular coffee drinks.     

Current-Mode Temperature Compensated Continuous-Time CMOS Transconductance-C Filter

In this paper a low voltage low power current-mode continuous-timelow-pass filter is presented. The current-integrator is constructedof a multi-output linearized transconductance element drivenby a balanced transimpedance amplifier. A 1 MHz third order ellipticlow pass filter is designed and fabricated with the proposedcurrent-mode integrator. The measurements show capability ofoperation with down to a 3 V supply voltage. The measured currentconsumption is 850 µA at room temperature andthe dynamic range is 66 dB. The distortion of the filter is below50 dBc with peak signal levels up to twice the quiescent currentlevel and with the refined designs even lower distortion is expected.The filter uses a CMOS-current reference for temperature andprocess variation stabilization with a measured corner frequencytemperature coefficient of –100 ppm/K.     

Volume integral equations in non-linear 3-D magnetostatics

In this paper a discussion of volume integral formulations in three-dimensional non-linear magnetostatics is presented. Integral formulations are examined in connection with Whitney's elements in order to find new approaches. A numerical algorithm based on a formulation implying properly the continuity conditions of magnetic field strength H, i.e. an h-type formulation, is introduced. Results of demanding application problems are shown demonstrating the characteristics of this kind of volume integral approach. In addition, a discussion of the parallelized version of the numerical code based on the h-type approach is presented appended with numerical results illustrating the advantages of combining integral formulations with concurrent computing.     

Diagnosing diabetic nephropathy by 1H NMR metabonomics of serum

Object: The most severe complication of type 1 diabetes (T1DM) is diabetic nephropathy. It is associated with a high risk of cardiovascular complications and premature death and requires early detection to be efficiently treated. The clinical practice to diagnose diabetic nephropathy is also a non-optimal and tedious set up based on albumin excretion rate in multiple overnight or 24h urine samples. Conversely, in this study, these independent diagnostic data are used to provide a realistic testing case for applying ¹H NMR metabonomics of serum in a diagnostic fashion. Materials and Methods: 182 T1DM and 21 non-diabetic (non-T1DM) individuals were studied. The ¹H NMR of serum at 500 MHz was targeted at two molecular windows: lipoprotein lipids and low-molecular-weight metabolites. Results: T1DM and non-T1DM individuals were exclusively separated by ¹H NMR. For diabetic nephropathy diagnosis in the T1DM patients, ¹H NMR data (and clinical biochemistry data) gave a sensitivity of 87.1% (83.9%) and a specificity of 87.7% (95.9%). The predictive values of positive and negative tests were 89.0% (95.5%) and 83.6% (79.2%), respectively. Conclusions: ¹H NMR metabonomics clearly distinguishes metabolic characteristics of T1DM and appears approximately as good a means to diagnose diabetic nephropathy from serum as an advanced set of biochemical variables.     

Experimental verification of a method for estimating energy for domestic hot water production in a 2-stage district heating substation

In this paper we compare our estimate of energy consumption for domestic hot water production in a building with the measured value. The energy consumption for hot water production is estimated from the measured total power consumption. The estimation method was developed using computer simulations, and it is based on the assumption that hot water production causes rapid and detectable changes in power consumption. A comparison of our estimates with measurements indicates that the uncertainty in estimation of hot water energy consumption is ±10%. Thus, the estimate is comparable to class 3 energy meter measurements, which have an uncertainty of ±2-10%.     

Application of capillary electrophoresis to determine metal cations,anions, organic acids,and carbohydrates in some Pinot Noir red wines

Capillary electrophoresis was used for determination of inorganic alkali and alkali earth metal cations, inorganic anions, anions of organic acids, and carbohydrates in six Pinot Noir grape red wines. Spectrum analysis was applied to evaluate the possible chemical differences between the wines from different geographic origin. In addition, sensory evaluation was used to reveal differences between wines from the end user point of view. Both chemical analyses and the sensory evaluation revealed clear differences among the wine samples. The total concentrations of inorganic ions, carboxylate anions, and carbohydrates varied 1.9–3.1 g L⁻¹, 4.3–5.5 g L⁻¹, and 1.5–7.5 g L⁻¹, respectively. Although the grape was same in all samples, the results show considerable variations in chemical composition, the red wine from Tasmanian being the most different among samples.