An empirical investigation of bias and variance in time series forecasting: modeling considerations and error evaluation

Bias and variance play an important role in understanding the fundamental issue of learning and generalization in neural network modeling. Several studies on bias and variance effects have been published in classification and regression related research of neural networks. However, little research has been done in this area for time-series modeling and forecasting. We consider modeling issues related to understanding error components given the common practices associated with neural-network time-series forecasting. We point out the key difference between classification and time-series problems in consideration of the bias-plus-variance decomposition. A Monte Carlo study on the role of bias and variance in neural networks time-series forecasting is conducted. We find that both input lag structure and hidden nodes are important in contributing to the overall forecasting performance. The results also suggest that overspecification of input nodes in neural network modeling does not impact the model bias, but has significant effect on the model variance. Methods such as neural ensembles that focus on reducing the model variance, therefore, can be valuable and effective in time-series forecasting modeling.     

Modulating drug release from poly(lactic-co-glycolic) acid microparticles by the addition of alginate and pectin

The aim of the present work is the characterization of PLGA microparticles including biopolymers for the controlled release of tilmicosin, a broad-spectrum antibiotic. Microparticles were prepared using the double-emulsion solvent evaporation technique. The effect of alginate and pectin incorporation over particle size and porosity, encapsulation efficiency (EE) and pH-responsive drug release was evaluated. Formulations presented a mean particle size of 5.5 μm approximately and a drug EE ranged from 22%–57%. PLGA-Alginate particles showed an increased porosity. Tilmicosin release profiles from PLGA and PLGA-biopolymer microparticles were affected by the particular combination of polymers and the pH of the release medium. The experimental data was simulated using a mathematical model, which takes into account the autocatalytic polymer degradation and the different mechanisms of drug transport. The combination of PLGA and biopolymers strongly influenced the morphology of the particles, offering the possibility of controlling the drug release profiles according to the therapy.     

Controlling Mechanical Behavior of TWIP Steels by Tuning Texture and Stacking Faults

Metallurgical and Materials Transactions A - We developed an Fe–22Mn–0.6C–1.5Al TWIP steel and investigated how thermo-mechanical processes affect the mechanical properties. In...     

Chemical Composition and Physical Properties of High Oleic Safflower Oils (Carthamus tinctorius,Var. CW88-OL and CW99-OL)

Chemical composition and physical properties of CW88‐OL and CW99‐OL cultivars of high oleic safflower seeds and their hexane‐extracted oils were determined. Dry‐based seed composition of CW88‐OL and CW99‐OL was: moisture = 4.29 and 4.23 %, oil = 42.29 and 46.44 %, Crude protein = 20.94 and 16.41 %, neutral detergent fiber = 28.11 and 28.49 %, ash = 1.55 and 2.01 %, phosphorus content = 2033 and 3995 mg/kg, respectively. Major fatty acids in oils were ~78 % oleic (O), ~13 % linoleic (L), ~5 % palmitic (P) and ~2 % stearic (St) acids, for both cultivars. The main triacylglycerols were OOO (~50 %), OOL (~20 %), SOL + OPO (~10 %), and LLP (~5 %). The oil composition of CW88‐OL and CW99‐OL in main minor components was: α‐tocopherol = 582 and 551 mg/kg, total sterols = 3996 and 3362 mg/kg, phospholipids = 22 and 21 mg/kg and wax content = 70 and 74 mg/kg. For both cultivars, density and viscosity of the oils between 25 and 55 °C varied from 903.4 to 912.6 kg/m³ and 63 to 23 mPa.s showing linear and exponential behaviors, respectively. The refractive index was 1.4694. The CIELab color parameters were: 89.69 and 89.53 (L*), ?3.72 and ?3.07 (a*), and 47.28 and 47.78 (b*) (CW88‐OL and CW99‐OL, respectively). Thus, the high oil content of the seeds and nutritional quality of the oil accompanied by low levels of waxes and phospholipids makes the cultivars studied promising for producers and consumers.     

A Fluorometric Enzymatic Assay for Quantification of Steryl Glucosides in Biodiesel

Steryl glucosides (SG) are common contaminants in biodiesel that form precipitates, which form and cause problems due to fouling during transport and storage. Therefore, their quantification is necessary to assess the quality of this fuel. The methods currently available for SG analysis require expensive instrumentation, need a previous concentration step by solid‐phase extraction (SPE) or are of limited use for the quantitative assessment. We developed an enzymatic method for SG quantification in biodiesel samples based on the hydrolysis of the glucoside catalyzed by a broadly specific beta glucosidase and the subsequent determination of the glucose released by the reaction. The method is non‐expensive, sensitive and was adapted to 96‐well format fluorescence plate reader, making it useful for the parallel assay of multiple samples. The enzymatic assay presented here represent a valuable tool for both quality control and the development of improved biodiesel production and purification procedures.     

Sirtuin 7 Deficiency Reduces Inflammation and Tubular Damage Induced by an Episode of Acute Kidney Injury

Acute kidney injury (AKI) is a public health problem worldwide. Sirtuins are a family of seven NAD+-dependent deacylases, Overexpression of Sirtuin 1, 3, and 5 protect against AKI. However, the role of Sirtuin 7 (Sirt7) in AKI is not known. Here, we analyzed how Sirt7 deficient mice (KO-Sirt7) were affected by AKI. As expected, wild-type and Sirt7 heterozygotes mice that underwent renal ischemia/reperfusion (IR) exhibited the characteristic hallmarks of AKI: renal dysfunction, tubular damage, albuminuria, increased oxidative stress, and renal inflammation. In contrast, the KO-Sirt7+IR mice were protected from AKI, exhibiting lesser albuminuria and reduction in urinary biomarkers of tubular damage, despite similar renal dysfunction. The renoprotection in the Sirt7-KO+IR group was associated with reduced kidney weight, minor expression of inflammatory cytokines and less renal infiltration of inflammatory cells. This anti-inflammatory effect was related to diminished p65 expression and in its active phosphorylation, as well as by a reduction in p65 nuclear translocation. Sirt7 deficient mice are protected from AKI, suggesting that this histone deacetylase promotes tubular damage and renal inflammation. Therefore, our findings indicate that Sirt7 inhibitors may be an attractive therapeutic target to reduce NFκB signaling.     

Interaction of the σ2 Receptor Ligand PB28 with the Human Nucleosome: Computational and Experimental Probes of Interaction with the H2A/H2B Dimer

Sigma‐2 (σ₂) binding sites are an emerging target for anti‐neoplastic agents due to the strong apoptotic effect exhibited by σ₂ agonists in vitro and the overexpression of these sites in tumor cells. Nonetheless, no σ₂ receptor protein has been identified. Affinity chromatography using the high‐affinity σ₂ ligand PB28 and human SK‐N‐SH neuroblastoma cells was previously utilized to identify σ₂ ligand binding proteins, specifically histones H1, H2A, H2B, and H3.3a. To rationalize this finding, homology modeling and automated docking studies were employed to probe intermolecular interactions between PB28 and human nucleosomal proteins. These studies predicted interaction of PB28 with the H2A/H2B dimer at a series of sites previously found to be implicated in chromatin compaction and nucleosomal assembly. To experimentally verify this prediction, a competitive binding assay was performed on the reconstituted H2A/H2B dimer using [³H]PB28 as radioligand, and an IC₅₀ value of 0.50 nM was determined for PB28 binding. In addition, [³H]PB28 was found to accumulate with up to a fivefold excess in nuclear fractions over cytosolic fractions of SK‐N‐SH and MCF7 cells, indicating that PB28 is capable of entering the nucleus to interact with histone proteins. In conjunction with computational results, these data suggest that PB28 may exert its cytotoxic effect through direct interaction with nuclear material.