NREMT Practice Questions
HEAL100 · Other
HEAL100 Health Care — exam and revision material · Other
Updated today
Health Care exam material, 23 pages. Originally published on StudyMerit.
See the first 3 pages exactly as they appear in the document.
Isye 6501 Final exam Questions and answers | Updated 2024/25 1-norm ii - ii ii ii Similar ii to ii rectilinear ii distance; ii measures ii the ii straight-line ii length ii of ii a ii vector ii from ii the ii origin. ii If ii z=(z1,z2,...,zm) ii is ii a ii vector ii in ii an ii m-dimensional ii space, ii then ii it's ii 1-norm ii is ii square ii root(| 𝑧 1|+| 𝑧 2|+ ⋯ +| 𝑧𝑚 | ii = ii | 𝑧 1|+| 𝑧 2|+ ⋯ +| 𝑧 | ii = ii Σm ii over ii i=1 ii | 𝑧𝑖 | ii A/B ii Testing ii - ii ii ii testing ii two ii alternatives ii to ii see ii which ii one ii performs ii better ii 2-norm ii - ii ii ii Similar ii to ii Euclidian ii distance; ii measures ii the ii straight-line ii length ii of ii a ii vector ii from ii the ii origin. ii If ii z=(z1,z2,...,zm) ii is ii a ii vector ii in ii an ii 𝑚 -dimensional ii space, ii then ii its ii 2-norm ii is ii the ii same ii as ii 1-norm ii but ii everything ii is ii squared= ii square ii root(Σm ii over ii i=1 ii (| 𝑧𝑖 |)^2) ii Accuracy ii - ii ii ii Fraction ii of ii data ii points ii correctly ii classified ii by ii a ii model; ii equal ii to ii TP+TN ii / ii TP+FP+TN+FN ii Action ii - ii ii ii In ii ARENA, ii something ii that ii is ii done ii to ii an ii entity. ii Additive ii Seasonality ii - ii ii ii Seasonal ii effect ii that ii is ii added ii to ii a ii baseline ii value ii (for ii example, ii "the ii temperature ii in ii June ii is ii 10 ii degrees ii above ii the ii annual ii baseline"). ii Adjusted ii R-squared ii - ii ii ii Variant ii of ii R2 ii that ii encourages ii simpler ii models ii by ii penalizing ii the ii use ii of ii too ii many ii variables. ii AIC ii - ii ii ii Akaike ii information ii criterion- ii Model ii selection ii technique ii that ii trades ii off ii between ii model ii fit ii and ii model ii complexity. ii When ii comparing ii models, ii the ii model ii with ii lower ii AIC ii is ii preferred. ii Generally ii penalizes ii complexity ii less ii than ii BIC. ii Algorithm ii - ii ii ii Step-by-step ii procedure ii designed ii to ii carry ii out ii a ii task. ii Analysis ii of ii Variance/ANOVA ii - ii ii ii Statistical ii method ii for ii dividing ii the ii variation ii in ii observations ii among ii different ii sources. ii Approximate ii dynamic ii program ii - ii ii ii Dynamic ii programming ii model ii where ii the ii value ii functions ii are ii approximated. ii Arc ii - ii ii ii Connection ii between ii two ii nodes/vertices ii in ii a ii network. ii In ii a ii network ii model, ii there ii is ii a ii variable ii for ii each ii arc, ii equal ii to ii the ii amount ii of ii flow ii on ii the ii arc, ii and ii (optionally) ii a ii capacity ii constraint ii on ii the ii arc's ii flow. ii Also ii called ii an ii edge. ii Area ii under ii the ii curve ii (AUC) ii - ii ii ii Area ii under ii the ii ROC ii curve; ii an ii estimate ii of ii the ii classification ii model's ii accuracy. ii Also ii called ii concordance ii index. ii ARIMA ii - ii ii ii Autoregressive ii integrated ii moving ii average.
ii Arrival ii Rate ii - ii ii ii Expected ii number ii of ii arrivals ii of ii people, ii things, ii etc. ii per ii unit ii time ii -- ii for ii example, ii the ii expected ii number ii of ii truck ii deliveries ii per ii hour ii to ii a ii warehouse. ii Assignment ii Problem ii - ii ii ii Network ii optimization ii model ii with ii two ii sets ii of ii nodes, ii that ii finds ii the ii best ii way ii to ii assign ii each ii node ii in ii one ii set ii to ii each ii node ii in ii the ii other ii set. ii Attribute ii - ii ii ii A ii characteristic ii or ii measurement ii - ii for ii example, ii a ii person's ii height ii or ii the ii color ii of ii a ii car. ii Generally ii interchangeable ii with ii "feature", ii and ii often ii with ii "covariate" ii or ii "predictor". ii In ii the ii standard ii tabular ii format, ii a ii column ii of ii data. ii Autoregression ii - ii ii ii Regression ii technique ii using ii past ii values ii of ii time ii series ii data ii as ii predictors ii of ii future ii values. ii Autoregressive ii integrated ii moving ii average ii (ARIMA) ii - ii ii ii Time ii series ii model ii that ii uses ii differences ii between ii observations ii when ii data ii is ii nonstationary. ii Also ii called ii Box-Jenkins. ii Backward ii elimination ii - ii ii ii Variable ii selection ii process ii that ii starts ii with ii all ii variables ii and ii then ii iteratively ii removes ii the ii least-immediately-relevant ii variables ii from ii the ii model. ii Balanced ii Design ii - ii ii ii Set ii of ii combinations ii of ii factor ii values ii across ii multiple ii factors, ii that ii has ii the ii same ii number ii of ii runs ii for ii all ii combinations ii of ii levels ii of ii one ii or ii more ii factors. ii Balking ii - ii ii ii An ii entity ii arrives ii to ii the ii queue, ii sees ii the ii size ii of ii the ii line ii (or ii some ii other ii attribute), ii and ii decides ii to ii leave ii the ii system. ii Bayes' ii theorem/Bayes' ii rule ii - ii ii ii Fundamental ii rule ii of ii conditional ii probability: ii 𝑃 ( 𝐴 | 𝐵 )= 𝑃 ( 𝐵 | 𝐴 )* 𝑃 ( 𝐴 ) ii / ii 𝑃 ( 𝐵 ) ii Bayesian ii Information ii criterion ii (BIC) ii - ii ii ii Model ii selection ii technique ii that ii trades ii off ii model ii fit ii and ii model ii complexity. ii When ii comparing ii models, ii the ii model ii with ii lower ii BIC ii is ii preferred. ii Generally ii penalizes ii complexity ii more ii than ii AIC. ii Bayesian ii Regression ii - ii ii ii Regression ii model ii that ii incorporates ii estimates ii of ii how ii coefficients ii and ii error ii are ii distributed. ii Bellman's ii Equation ii - ii ii ii Equation ii used ii in ii dynamic ii programming ii that ii ensures ii optimality ii of ii a ii solution. ii Bernoulli ii Distribution ii - ii ii ii Discrete ii probability ii distribution ii where ii the ii outcome ii is ii binary, ii either ii 0 ii or ii 1. ii Often, ii 1 ii represents ii success ii and ii 0 ii represents ii failure. ii The ii probability ii of ii the ii outcome ii being ii 1 ii is ii 𝑝 ii and ii the ii probability ii of ii outcome ii being ii 0 ii is ii 𝑞 ii = ii 1− 𝑝 , ii where ii 𝑝 ii is ii between ii 0 ii and ii 1. ii Bias ii - ii ii ii Systematic ii difference ii between ii a ii true ii parameter ii of ii a ii population ii and ii its ii estimate.
ii Binary ii Data ii - ii ii ii Data ii that ii can ii take ii only ii two ii different ii values ii (true/false, ii 0/1, ii black/white, ii on/off, ii etc.) ii Binary ii integer ii program ii - ii ii ii Integer ii program ii where ii all ii variables ii are ii binary ii variables. ii Binary ii Variable ii - ii ii ii Variable ii that ii can ii take ii just ii two ii values: ii 0 ii and ii 1. ii Binomial ii Distribution ii - ii ii ii Discrete ii probability ii distribution ii for ii the ii exact ii number ii of ii successes, ii k, ii out ii of ii a ii total ii of ii n ii iid ii Bernoulli ii trials, ii each ii with ii probability ii p: ii Pr( 𝑘 )= ii (n ii over ii k) ii p^k(1-p)^n-k ii Blocking ii - ii ii ii Factor ii introduced ii to ii an ii experimental ii design ii that ii interacts ii with ii the ii effect ii of ii the ii factors ii to ii be ii studied. ii The ii effect ii of ii the ii factors ii is ii studied ii within ii the ii same ii level ii (block) ii of ii the ii blocking ii factor. ii box ii and ii whisker ii plot ii - ii ii ii Graphical ii representation ii data ii showing ii the ii middle ii range ii of ii data ii (the ii "box"), ii reasonable ii ranges ii of ii variability ii ("whiskers"), ii and ii points ii (possible ii outliers) ii outside ii those ii ranges. ii Box-Cox ii Transformation ii - ii ii ii Transformation ii of ii a ii non-normally-distributed ii response ii to ii a ii normal ii distribution. ii Branching ii - ii ii ii Splitting ii a ii set ii of ii data ii into ii two ii or ii more ii subsets, ii to ii each ii be ii analyzed ii separately. ii CART ii - ii ii ii Classification ii and ii regression ii trees. ii Categorical ii Data ii - ii ii ii Data ii that ii classifies ii observations ii without ii quantitative ii meaning ii (for ii example, ii colors ii of ii cars) ii or ii where ii quantitative ii amounts ii are ii categorized ii (for ii example, ii "0- 10, ii 11-20, ii ..."). ii Causation ii - ii ii ii Relationship ii in ii which ii one ii thing ii makes ii another ii happen ii (i.e., ii one ii thing ii causes ii another). ii Chance ii Constraint ii - ii ii ii A ii probability-based ii constraint. ii For ii example, ii a ii standard ii linear ii constraint ii might ii be ii 𝐴 x≤ 𝑏 . ii A ii similar ii chance ii constraint ii might ii be ii Pr ii ( 𝐴 x≤ 𝑏 )≥0.95 ii Change ii Detection ii - ii ii ii Identifying ii when ii a ii significant ii change ii has ii taken ii place ii in ii a ii process. ii Classification ii - ii ii ii The ii separation ii of ii data ii into ii two ii or ii more ii categories, ii or ii (a ii point's ii classification) ii the ii category ii a ii data ii point ii is ii put ii into. ii Classification ii tree ii - ii ii ii Tree-based ii method ii for ii classification. ii After ii branching ii to ii split ii the ii data, ii each ii subset ii is ii analyzed ii with ii its ii own ii classification ii model. ii Classifier ii - ii ii ii A ii boundary ii that ii separates ii the ii data ii into ii two ii or ii more ii categories. ii Also ii (more ii generally) ii an ii algorithm ii that ii performs ii classification.
20 more pages after this
You’ve read 3 of 23. Unlock once and it stays in your library — and if the rest doesn’t match what you just read, you get a refund.
HEAL100 · Other
D218 · Western Governors University
HEAL100 · Other
$7.50
Yours permanently