scholarly journals On the Origin of the Standardization Sensitivity in RegEM Climate Field Reconstructions*

2008 ◽  
Vol 21 (24) ◽  
pp. 6710-6723 ◽  
Author(s):  
Jason E. Smerdon ◽  
Alexey Kaplan ◽  
Diana Chang
Keyword(s):  
Standard Deviation ◽  
Missing Values ◽  
Mean Field ◽  
Alternative Methods ◽  
Reconstruction Method ◽  
Climate Data ◽  
The Mean ◽  
Calibration Interval ◽  
Reconstruction Interval ◽  
Selection Of

Abstract The regularized expectation maximization (RegEM) method has been used in recent studies to derive climate field reconstructions of Northern Hemisphere temperatures during the last millennium. Original pseudoproxy experiments that tested RegEM [with ridge regression regularization (RegEM-Ridge)] standardized the input data in a way that improved the performance of the reconstruction method, but included data from the reconstruction interval for estimates of the mean and standard deviation of the climate field—information that is not available in real-world reconstruction problems. When standardizations are confined to the calibration interval only, pseudoproxy reconstructions performed with RegEM-Ridge suffer from warm biases and variance losses. Only cursory explanations of this so-called standardization sensitivity of RegEM-Ridge have been published, but they have suggested that the selection of the regularization (ridge) parameter by means of minimizing the generalized cross validation (GCV) function is the source of the effect. The origin of the standardization sensitivity is more thoroughly investigated herein and is shown not to be associated with the selection of the ridge parameter; sets of derived reconstructions reveal that GCV-selected ridge parameters are minimally different for reconstructions standardized either over both the reconstruction and calibration interval or over the calibration interval only. While GCV may select ridge parameters that are different from those that precisely minimize the error in pseudoproxy reconstructions, RegEM reconstructions performed with truly optimized ridge parameters are not significantly different from those that use GCV-selected ridge parameters. The true source of the standardization sensitivity is attributable to the inclusion or exclusion of additional information provided by the reconstruction interval, namely, the mean and standard deviation fields computed for the complete modeled dataset. These fields are significantly different from those for the calibration period alone because of the violation of a standard EM assumption that missing values are missing at random in typical paleoreconstruction problems; climate data are predominantly missing in the preinstrumental period when the mean climate was significantly colder than the mean of the instrumental period. The origin of the standardization sensitivity therefore is not associated specifically with RegEM-Ridge, and more recent attempts to regularize the EM algorithm using truncated total least squares could theoretically also be susceptible to the problems affecting RegEM-Ridge. Nevertheless, the principal failure of RegEM-Ridge arises because of a poor initial estimate of the mean field, and therefore leaves open the possibility that alternative methods may perform better.

SPECIAL ARTICLE

PEDIATRICS ◽  
1954 ◽  
Vol 14 (5) ◽  
pp. 563-564
Author(s):  
Keyword(s):  
Standard Deviation ◽  
Visual Assessment ◽  
Probable Error ◽  
Oral Examination ◽  
Frequency Curve ◽  
Written Examination ◽  
Lower Figure ◽  
Normal Frequency ◽  
The Mean ◽  
Selection Of

THE WRITTEN examination of January 15, 1954, was taken by 515 candidates, a larger number than in any previous year except 1953 when there were 607 candidates Grades ranged from a lowest mark of 32.0 to a highest mark of 89.5 Inspection of the range resulted in the decision to place the passing mark at 51. On this basis there were 32, or 6.2%, who failed and were therefore ineligible for oral examination. The distribution of the grades earned by the 515 candidates is presented in the form of a histogram. As an aid to visual assessment of the nature of the distribution, a normal frequency curve computed from the mean and standard deviation of the data has been superimposed on the diagram. The distribution of the grades is clearly and impressively skewed to the left, that is, the scores tend to be massed at the high end of the scale and spread out at the low end. An examination of this type is relatively sensitive in the zone of poorer scholarship where the selection of failures is to be made and relatively insensitive in the range of higher scholarship. The intrinsic reliability of the examination has again been assessed by comparing the grade made by each candidate on his odd-numbered questions with that earned on his even-numbered questions. The comparison reveals a "probable error of estimate," P.E.m, of 2.39, a lower figure and therefore a higher degree of reliability than yielded by any previous examination. The Committee is pleased that the increase in reliability was accomplished in spite of the fact that the length of the examination was decreased from 250 grading points in 1953 to 200 grading points.

scholarly journals Revisiting the functional response in habitat selection for large herbivores: a matter of spatial variation in resource distribution?

2019 ◽  
Vol 30 (6) ◽  
pp. 1725-1733 ◽  
Author(s):  
Antoine Duparc ◽  
Mathieu Garel ◽  
Pascal Marchand ◽  
Dominique Dubray ◽  
Daniel Maillard ◽  
...  
Keyword(s):  
Standard Deviation ◽  
Habitat Selection ◽  
Spatial Variation ◽  
Functional Response ◽  
Resource Distribution ◽  
Habitat Characteristics ◽  
High Rate ◽  
Large Herbivores ◽  
The Mean ◽  
Selection Of

Abstract Most habitats are distributed heterogeneously in space, forcing animals to move according to both habitat characteristics and their needs for energy and safety. Animal space use should therefore vary according to habitat characteristics, a process known as the “functional response” in habitat selection. This response has often been tested vis-à-vis the proportion of a habitat category within areas available to individuals. Measuring sought-after resources in landscape where they are continuously distributed is a challenge and we posit here that both the mean availability of a resource and its spatial variation should be measured. Accordingly, we tested for a functional response in habitat selection according to these two descriptors of the resource available for a mountain herbivore. We hypothesized that selection should decrease with mean value of resources available and increase with its spatial variation. Based on GPS data from 50 chamois females and data on the actual foodscape (i.e., distribution of edible-only biomass in the landscape), we estimated individual selection ratio (during summer months) for biomass at the home range level, comparing edible biomass in individual home ranges and the mean and standard deviation of edible biomass in their available range. Chamois being a group-living species, available accessible ranges were shared by several individuals that formed socio-spatial groups (clusters) in the population. As expected, selection ratios increased with the standard deviation of edible resources in each cluster, but unlike our prediction, was unrelated to its mean. Selection of areas richer in resources hence did not fade away when more resources were available on average, a result that may be explained by the need for this capital breeder species to accumulate fat-reserve at a high rate during summer months. Low spatial variation could limit the selection of chamois, which highlights the importance of resource distribution in the process of habitat selection.

scholarly journals Correlation between Personality Types and Color Shade Preference

2014 ◽  
Vol 1 (4) ◽  
Author(s):  
Divya Ghorawat ◽  
Ravina Madan
Keyword(s):  
Standard Deviation ◽  
Preference Test ◽  
Personality Types ◽  
Personality Test ◽  
Age Group ◽  
Colour Preference ◽  
Individual Personality ◽  
The Mean ◽  
The Difference ◽  
Selection Of

The main aim of this study was to find out whether the color shade preferences of individuals, grouped according to their personality types are alike or not. The researchers also tried to study the difference in color shade preferences based on gender differences. The objective was to understand whether the color shade preferences of individuals are affected by their individual personality types or not and whether gender plays any role in selection of color shades of individuals. The study sample consisted of a total of 80 subjects, randomly selected from within the age group of 18-25. An attempt was made to maintain a balance between two genders and across the ages. The research instruments used were Eysenck’s Personality Test (to measure the personality types) and a separate Colour Bar- Colour Preference Test was designed to serve the purpose of this study. The subjects were made to fill both the questionnaires and then the items were scored and results were analyzed. Positive scoring was done for Eysenck’s personality questionnaire and negative scoring was used to score the Colour Bar-Colour Preference Test. The data collected was divided into: Introverts (female, male) and Extroverts (female, male). From the data collected the mean, standard deviation and correlation between extraversion scores and the most preferred colour scores, for all the four groups. The mean, standard deviation, and correlation for the four groups was found to be 436.4 (IF), 529.2 (IM), 504 (EF), 453.4 (EM); 5.47 (IF), 6.52 (IM), 5.48 (EF), 6.93(EM); -0.3 (IF), -0.1 (IM), 0.1 (EF) and 0.5 (EM) respectively. From this the conclusion was drawn that there was no significant correlation found between the personality types and colour shade preferences of the subjects except in the case of extroverted males. Recommendations: Increase the sample size as that proved to be a limitation to this study. A study across ages is also possible and might give different results.

scholarly journals Biases in galaxy cluster velocity dispersion and mass estimates in the small Ngal regime

2020 ◽  
Vol 641 ◽  
pp. A41 ◽  
Author(s):  
A. Ferragamo ◽  
J. A. Rubiño-Martín ◽  
J. Betancort-Rijo ◽  
E. Munari ◽  
B. Sartoris ◽  
...  
Keyword(s):  
Standard Deviation ◽  
Velocity Dispersion ◽  
Galaxy Cluster ◽  
Mean Velocity ◽  
Massive Galaxies ◽  
Unbiased Estimators ◽  
Dispersion Estimate ◽  
Physical Effects ◽  
The Mean ◽  
Selection Of

Aims. We present a study of the statistical properties of three velocity dispersion and mass estimators: biweight, gapper, and standard deviation for a small number of galaxies (Ngal ≤ 75). Methods. Using a set of 73 numerically simulated galaxy clusters, we first characterised the statistical bias and the variance for each one of the three estimators (biweight, gapper, and standard deviation) in the determination of the velocity dispersion and the dynamical mass of the clusters through the σ–M relation. These results were used to define a new set of unbiased estimators that are able to correct for these statistical biases with a minimum increase in associated variance. We also used the same set of numerical simulations to characterise two other physical biases that affect the estimates: the effect of velocity segregation on the selection of cluster members, and the effect of using cluster members within different physical radii from the cluster centre. Results. The standard deviation (and its unbiased counterpart) is the estimator with the lowest variance estimator after the biweight and gapper. The effect of velocity segregation in the selection of galaxies within the sub-sample of the most massive galaxies in the cluster introduces a bias of 2% in the velocity dispersion estimate when it is calculated using a quarter of the most massive cluster members. We also find a dependence of the velocity dispersion estimate on the aperture radius as a fraction of R200. This is consistent with previous results in the literature. Conclusions. The proposed set of unbiased estimators effectively provides a correction of the velocity dispersion and mass estimates from the statistical and physical effects discussed above for small numbers of cluster members. When these new estimators are applied to a subset of simulated observations, they can retrieve bias-corrected values for the mean velocity dispersion and the mean mass; the standard deviation has the lowest variance. Although for a single galaxy cluster the statistical and physical effects discussed here are comparable to or slightly smaller than the bias introduced by interlopers, they are relevant when ensemble properties and scaling relations for large number of clusters are studied.

Research on twin-SIR rumor spreading model in online social network

2020 ◽  
pp. 1-12
Author(s):  
Jing Yi ◽  
Peiyu Liu ◽  
Zhihao Wang ◽  
Wenfeng Liu
Keyword(s):  
Online Social Network ◽  
Mean Field ◽  
Traditional Media ◽  
Spreading Process ◽  
Rumor Spreading ◽  
News Reports ◽  
Spreading Model ◽  
The Mean ◽  
Mean Field Equation ◽  
Selection Of

In the study filed of rumor spreading, kill rumor or dispel rumor is very important in order to control rumor spreading and reduce the bad influence of the rumor. In the previous studies, rumor clarification is mostly finished by relying on external media or news reports instead of intervening and controlling from inside the network, which causes that the speed of rumor clarification is far lower than the speed of rumor spreading, and it is not ideal for the effect of rumor clarification. In this paper, a new Twin-SIR spreading model is proposed, in which, a rumor clarification node named as “rumor dispeller” with the spreading ability is introduced. The rumor dispeller is involved in the spreading process of the model together with the rumor spreader to control the spreading of rumor and thus to achieve the purpose of clarifying rumor. At the same time, during the process of building the model, we also apply the traditional media as a spreading parameter to the spreading process of the model. We built the mean-field equation of the model and then implemented further analysis of the model on homogeneous networks and heterogeneous networks. Through experimental simulations, the “rumor dispeller” was found to have the ability to reduce the spread of rumor spreading, and that the selection of the initial “rumor dispeller” node can affect the effect of rumor spreading, and at the same time, the external media have an important influence on rumor clarification. These conclusions have a new function for guiding us to study the mechanism of rumor spreading.

367. Bacterial growth in reconstituted spray-dried milk

1948 ◽  
Vol 15 (3) ◽  
pp. 285-291 ◽  
Author(s):  
Constance Higginbottom
Keyword(s):  
Standard Deviation ◽  
Bacterial Growth ◽  
Food Poisoning ◽  
Spore Count ◽  
Fresh Milk ◽  
Standard Deviation Range ◽  
The Mean ◽  
Reconstituted Milk ◽  
Spray Dried ◽  
Selection Of

1. Over 200 samples of spray-dried milks from eight British plants have been examined for total and spore counts, numbers of β-haemolytic colonies and for moulds, yeasts and conliform bacteria.2. The mean count and standard deviation range for total and spore count are reported. No β-haemolytic colonies other than those due to spore-forming bacilli were detected.3. After ageing the reconstituted milk for 24 hr. at 15–5° C. the mean count (37° C.) was unchanged; at 22° C. the increase was 2000-fold and at 37° C. the milk had clotted within this time.4. The predominating organisms in a random selection of the reconstituted milks are described, both for the fresh milk and after ageing at 15·5 and 22° C. A marked increase i n spore-forming bacilli was noted at 22° C.5. The relation of the flora of reconstituted milk to its food-poisoning potentialities is briefly discussed.

EVALUATION OF PROCEDURES FOR ESTIMATING WITHIN-SPOT POPULATIONS OF ATTACKING ADULT DENDROCTONUS FRONTALIS (COLEOPTERA: SCOLYTIDAE)

1977 ◽  
Vol 109 (10) ◽  
pp. 1325-1334 ◽  
Author(s):  
Paul E. Pulley ◽  
John L. Foltz ◽  
Robert N. Coulson ◽  
William C. Martin
Keyword(s):  
Standard Deviation ◽  
Adult Stage ◽  
Population Data ◽  
Dendroctonus Frontalis ◽  
Single Level ◽  
Functional Relationships ◽  
True Number ◽  
Unbiased Estimates ◽  
The Mean ◽  
Selection Of

AbstractPopulation data collected from 132 trees during a 3-year period were used to simulate spots of K trees (1≤ k ≤ 50) infested by the attacking adult stage of Dendroctonus frontalis Zimmerman. The total number of beetles on the K trees was then estimated by sampling k trees (1≤ k ≤ 10). The k trees were chosen at random and by selecting those of largest diameter and largest infested area. Within-tree populations were estimated at two levels of precision and within-spot populations were then estimated by scaling the sum of the k within-tree estimates according to the proportion of the tree numbers, tree diameters, or infested phloem areas included in the sample. The various combinations of tree selection, within-tree precision, and scaling produced 10 procedures which were evaluated for bias, precision, and cost as estimators of within-spot populations. Bias was calculated as the mean of the proportional errors in estimating the true numbers, and the standard deviation of the proportional errors was used as a measure of precision.The procedures in which trees were randomly selected provided unbiased estimates of the within-spot populations. Selecting the largest trees tended to overestimate the true number with the bias diminishing to zero as k → K. However, separate analyses of trees sampled on the same date within actual spots showed no reason to reject the hypothesis of no difference in beetle density (insects/diameter and insects/area) between the largest and smallest trees.When k = K = 1, the precision of all within-spot estimators was equivalent to the precision of the within-tree estimate. For larger k = K, the precision improved approximately as √(K). No attempt was made to derive functional relationships of precision for k < K. For each procedure, precision improved as k → K. Sampling the k trees at two sample heights (3.5 and 6.5 m, 4–100 cm2 disks/height) was more precise than single level sampling (4 disks at 5 m), but equally precise estimates could be obtained by single level sampling of just one or two additional trees in the spot. Random selection of the k trees with scaling by the number of infested trees was the least precise of the estimating procedures; scaling by diameter and by infested surface area increased the precision. Best precision was obtained by selecting the k trees of greatest infested phloem area, but selecting the largest diameter trees was nearly as precise. The least costly procedure for obtaining a desired level of precision consists of selecting the k trees of largest diameter and extracting 4 disks/tree at 5 m.

scholarly journals Operational Risk Reverse Stress Testing: Optimal Solutions

2021 ◽  
Vol 26 (2) ◽  
pp. 38
Author(s):  
Peter Mitic
Keyword(s):  
Standard Deviation ◽  
Bayesian Methods ◽  
Optimization Problem ◽  
Stress Testing ◽  
Alternative Methods ◽  
Stochastic Algorithm ◽  
Simple Modification ◽  
Optimal Value ◽  
The Mean ◽  
Stop Condition

Selecting a suitable method to solve a black-box optimization problem that uses noisy data was considered. A targeted stop condition for the function to be optimized, implemented as a stochastic algorithm, makes established Bayesian methods inadmissible. A simple modification was proposed and shown to improve optimization the efficiency considerably. The optimization effectiveness was measured in terms of the mean and standard deviation of the number of function evaluations required to achieve the target. Comparisons with alternative methods showed that the modified Bayesian method and binary search were both performant, but in different ways. In a sequence of identical runs, the former had a lower expected value for the number of runs needed to find an optimal value. The latter had a lower standard deviation for the same sequence of runs. Additionally, we suggested a way to find an approximate solution to the same problem using symbolic computation. Faster results could be obtained at the expense of some impaired accuracy and increased memory requirements.

Hand anthropometry survey in Iranian adults and comparisons with other populations

Work ◽  
2021 ◽  
pp. 1-12
Author(s):  
Mohammad Hajaghazadeh ◽  
Mohsen Taghizadeh ◽  
Hamidreza Khalkhali ◽  
Iraj Mohebbi
Keyword(s):  
Standard Deviation ◽  
Iranian Population ◽  
Published Data ◽  
Hand Tools ◽  
Male And Female ◽  
Right Hand ◽  
The Mean ◽  
The Right ◽  
Digital Caliper ◽  
Selection Of

BACKGROUND: Hand anthropometry is useful for designing manual systems such as hand tools, controls, and gloves. There are limited published data on the hand dimensions of Iranian male and female adults. OBJECTIVE: This study was undertaken to measure the hand anthropometric dimensions in Iranian adults to compare data between two genders and also with the corresponding data from other nationalities. METHODS: A total of 34 dimensions of the right hand were measured in 217 male and 128 female adults using a digital caliper and tape. The mean, standard deviation, and the main percentiles are summarized in a table and the mean of each dimension is compared by independent t-tests between genders. Hand dimensions are compared between Iranian and other nationalities such as North Colombian, Korean, Turkish, Nigerian, Bangladeshi, Jordanian, and Vietnamese using the published data. RESULTS: Females had significantly smaller hand dimensions than males in all dimensions ranged from 4.21%to 18.16%, with the largest differences in hand breadths. Compared with other nationalities, the Iranian male and female adults had wider (greater breadth and circumferences) hands with shorter fingers. CONCLUSIONS: The results showed significant differences of hand anthropometry between Iranian and other nationalities which should be included in the design and selection of hand tools for Iranian population.

scholarly journals A Bayesian ANOVA Scheme for Calculating Climate Anomalies, with Applications to the Instrumental Temperature Record

2012 ◽  
Vol 25 (2) ◽  
pp. 777-791 ◽  
Author(s):  
Martin P. Tingley
Keyword(s):  
Time Series ◽  
Standard Deviation ◽  
Missing Values ◽  
Reference Interval ◽  
Temperature Record ◽  
Uncertainty Interval ◽  
Common Reference ◽  
Climate Anomalies ◽  
Spatial Moment ◽  
The Mean

Abstract Climate datasets with both spatial and temporal components are often studied after removing from each time series a temporal mean calculated over a common reference interval, which is generally shorter than the overall length of the dataset. The use of a short reference interval affects the temporal properties of the variability across the records, by reducing the standard deviation within the reference interval and inflating it elsewhere. For an annually averaged version of the Climate Research Unit’s (CRU) temperature anomaly product, the mean standard deviation is 0.67°C within the 1961–90 reference interval, and 0.81°C elsewhere. The calculation of anomalies can be interpreted in terms of a two-factor analysis of variance model. Within a Bayesian inference framework, any missing values are viewed as additional parameters, and the reference interval is specified as the full length of the dataset. This Bayesian scheme is used to re-express the CRU dataset as anomalies with respect to means calculated over the entire 1850–2009 interval spanned by the dataset. The mean standard deviation is increased to 0.69°C within the original 1961–90 reference interval, and reduced to 0.76°C elsewhere. The choice of reference interval thus has a predictable and demonstrable effect on the second spatial moment time series of the CRU dataset. The spatial mean time series is in this case largely unaffected: the amplitude of spatial mean temperature change is reduced by 0.1°C when using the 1850–2009 reference interval, while the 90% uncertainty interval of (−0.03, 0.23) indicates that the reduction is not statistically significant.

Sign in / Sign up

Export Citation Format

Share Document