Meta tags:
description= 15 posts published by Lars Syll during October 2016;
Headings (most frequently used words):
economics, the, and, in, what, of, is, mainstream, econometrics, science, be, svensk, universitetsutbildning, fritt, fall, to, recent, comments, lars, syll, some, unsettled, questions, macroeconomic, theory, rabbits, principle, triumph, ideology, over, non, ideological, valuefree, ll, dipped, truth, perils, calling, your, pet, cat, dog, ii, it, takes, make, real, statistical, significance, tests, do, not, validate, models, why, values, cannot, taken, at, face, value, economists, dissing, people, that, want, rethink, bridge, between, model, reality, on, proper, use, mathematics, posts, policy, reading, list, categories, archives,
Text of the page (most frequently used words):
the (262), and (146), that (129), not (59), are (55), #economics (46), this (40), with (36), for (36), och (35), but (35), you (34), have (33), oct (31), att (30), #models (29), 2016 (28), from (28), theory (26), what (26), one (26), they (24), when (24), model (23), there (23), data (22), may (21), jan (21), som (21), all (21), economic (20), can (20), real (20), till (20), world (19), science (18), econometrics (18), comments (17), more (17), statistical (17), our (17), mar (16), apr (16), jun (16), jul (16), posted (16), mainstream (16), med (16), aug (15), sep (15), nov (15), dec (15), feb (15), only (15), than (15), set (15), significance (15), har (15), 2014 (14), like (14), assumptions (14), want (14), new (14), economists (14), det (14), being (13), has (13), such (13), could (13), kind (13), population (13), many (13), 2012 (12), 2013 (12), 2015 (12), 2017 (12), 2018 (12), 2019 (12), 2020 (12), 2021 (12), 2022 (12), 2023 (12), 2024 (12), 2025 (12), statistics (12), most (12), some (12), should (12), hypothesis (12), test (12), its (12), inte (11), about (11), would (11), true (11), relevance (11), för (11), 2011 (10), reality (10), rather (10), krugman (10), studenter (10), time (9), how (9), truth (9), course (9), out (9), between (9), really (9), things (9), which (9), share (9), opens (9), window (9), standard (9), wrong (9), men (9), conclusions (9), lars (8), now (8), lönebildning (8), arbetslöshet (8), mathematical (8), probability (8), just (8), variables (8), based (8), these (8), because (8), problem (8), without (8), even (8), whether (8), their (8), scientific (8), your (8), content (7), other (7), 2026 (7), will (7), same (7), must (7), those (7), who (7), over (7), haavelmo (7), also (7), random (7), often (7), populations (7), them (7), rational (7), theories (7), where (7), result (7), women (7), thought (7), tests (7), universitet (7), detta (7), off (7), comment (6), arbetslöshetssiffran (6), styr (6), ett (6), den (6), right (6), way (6), give (6), possible (6), use (6), enough (6), level (6), modern (6), any (6), make (6), were (6), nothing (6), had (6), bluesky (6), facebook (6), students (6), people (6), neoclassical (6), don (6), economist (6), value (6), get (6), null (6), allt (6), många (6), högskolan (6), market (6), never (5), claim (5), think (5), better (5), indeed (5), believe (5), hope (5), causal (5), econometricians (5), probabilistic (5), cases (5), know (5), does (5), assumption (5), imaginary (5), used (5), why (5), larger (5), good (5), always (5), ask (5), case (5), see (5), least (5), relations (5), non (5), made (5), individuals (5), consider (5), say (5), researcher (5), perform (5), significant (5), then (5), two (5), context (5), among (5), under (5), important (5), fall (5), kan (5), svensk (5), universitetsutbildning (5), expectations (5), knowledge (5), minimum (5), write (4), view (4), syll (4), blog (4), fred (4), recent (4), especially (4), policy (4), research (4), precision (4), reasoning (4), relation (4), found (4), empirical (4), observations (4), stochastic (4), 1943 (4), values (4), itself (4), argue (4), approach (4), sample (4), samples (4), actually (4), very (4), observable (4), able (4), explanations (4), different (4), something (4), try (4), situations (4), years (4), today (4), yours (4), truly (4), analysis (4), everything (4), trying (4), search (4), reason (4), care (4), ideological (4), usually (4), sampling (4), first (4), threshold (4), best (4), effect (4), act (4), idag (4), från (4), hur (4), mer (4), dessa (4), man (4), sig (4), studenterna (4), fler (4), dessutom (4), fritt (4), rigor (4), questions (4), development (4), described (4), understand (4), false (4), wage (4), inferential (4), education (3), school (3), wiklund (3), milch (3), risksamhällets (3), konturer (3), follow (3), ones (3), limited (3), put (3), results (3), cannot (3), against (3), mathematics (3), sheer (3), accept (3), econometric (3), parameters (3), regularities (3), assume (3), social (3), find (3), theoretical (3), show (3), given (3), treat (3), anything (3), else (3), into (3), showing (3), additivity (3), investigator (3), principle (3), regression (3), david (3), experimental (3), working (3), axiomatic (3), difficult (3), turn (3), answer (3), treating (3), known (3), less (3), hand (3), however (3), little (3), bridge (3), processes (3), logically (3), unlikely (3), according (3), relative (3), explanatory (3), number (3), relevant (3), since (3), look (3), predictions (3), possibility (3), rendahl (3), mean (3), perfect (3), using (3), alternative (3), dipped (3), shown (3), lot (3), hard (3), totally (3), problems (3), reasonable (3), lead (3), draw (3), inferences (3), either (3), point (3), consistent (3), example (3), validate (3), coefficients (3), big (3), researchers (3), form (3), hypotheses (3), making (3), doesn (3), though (3), tool (3), need (3), depends (3), avoid (3), too (3), lunch (3), question (3), resurserna (3), studentgrupper (3), högskolor (3), går (3), mot (3), denna (3), tyvärr (3), sämre (3), rum (3), kraven (3), platser (3), kvalitet (3), ger (3), upp (3), kvaliteten (3), där (3), var (3), much (3), let (3), macroeconomic (3), theorists (3), seem (3), facts (3), related (3), work (3), study (3), fact (3), current (3), proposition (3), pedagogisk (3), könsmaktsordningen (3), rhythms (3), believes (3), doktorsavhandlingen (3), cat (3), dog (3), called (3), explain (3), simply (3), self (3), jersey (3), pennsylvania (3), unemployment (3), chicago (3), triumph (3), ideology (3), superpopulation (3), rabbit (3), required (2), log (2), sign (2), subscribed (2), subscribe (2), wordpress (2), com (2), account (2), methodology (2), society (2), reading (2), arbetslösheten (2), pseudo (2), irrelevant (2), sverige (2), causality (2), understood (2), far (2), next (2), elegance (2), expecting (2), calculation (2), comes (2), hypothetical (2), ought (2), argument (2), fallible (2), themselves (2), distribution (2), sad (2), open (2), system (2), equations (2), stable (2), invariant (2), exist (2), phenomena (2), mechanisms (2), underlying (2), his (2), metaphysical (2), uncover (2), convincing (2), expected (2), seen (2), take (2), leap (2), applying (2), left (2), studied (2), unlimited (2), merely (2), exists (2), ill (2), sense (2), further (2), treated (2), assumed (2), nevertheless (2), seems (2), free (2), necessity (2), understanding (2), inference (2), seriously (2), scientists (2), pure (2), systems (2), urge (2), infinite (2), collective (2), become (2), field (2), him (2), doing (2), easy (2), conclude (2), methods (2), supposed (2), business (2), before (2), apply (2), appropriate (2), trick (2), own (2), certain (2), law (2), hoping (2), build (2), essentially (2), 1944 (2), every (2), associated (2), particular (2), economies (2), terms (2), uncertainty (2), intrinsic (2), closure (2), great (2), invariance (2), limitations (2), complete (2), logical (2), finally (2), jump (2), testing (2), part (2), priori (2), sequences (2), somehow (2), typical (2), although (2), young (2), clearly (2), concept (2), equilibrium (2), nonsense (2), imagine (2), orthodoxy (2), basically (2), guy (2), perhaps (2), becomes (2), 1996 (2), couple (2), solving (2), guess (2), looking (2), disappointed (2), shows (2), ago (2), defend (2), teaching (2), textbooks (2), paul (2), after (2), compared (2), likely (2), incorrectly (2), democrats (2), republicans (2), health (2), military (2), matters (2), expect (2), pattern (2), explicable (2), idea (2), issue (2), political (2), corresponds (2), variable (2), face (2), takes (2), means (2), three (2), journal (2), intercept (2), several (2), stated (2), except (2), applied (2), tendency (2), disconfirmed (2), tells (2), error (2), disconfirm (2), reasons (2), lessons (2), human (2), concomitant (2), powerful (2), willing (2), percent (2), risks (2), place (2), she (2), kids (2), här (2), ytterligare (2), högskola (2), utbildningar (2), nya (2), större (2), utsträckning (2), ska (2), högre (2), femtio (2), vid (2), våra (2), befolkningen (2), antagande (2), höger (2), tar (2), lika (2), takt (2), vidare (2), ökar (2), får (2), över (2), sina (2), studier (2), situation (2), dem (2), eller (2), högskolestudier (2), skett (2), blivit (2), vara (2), politiskt (2), varför (2), samtidigt (2), visar (2), bland (2), finns (2), när (2), ifrågasatt (2), måste (2), genom (2), resurser (2), högskolorna (2), godkänt (2), annat (2), utbildning (2), robert (2), unrealistic (2), monetary (2), literature (2), influence (2), agree (2), definition (2), runs (2), systematic (2), criterion (2), through (2), jag (2), både (2), människa (2), häst (2), nummer (2), ways (2), power (2), while (2), seriality (2), contains (2), defenders (2), primary (2), järnvägsstationerna (2), reproducerar (2), met (2), invalid (2), calling (2), pet (2), tell (2), linearity (2), predictors (2), effects (2), aumann (2), tooth (2), organize (2), handy (2), valuefree (2), card (2), information (2), graduate (2), evidence (2), programmes (2), come (2), participants (2), perfectly (2), markets (2), invisible (2), news (2), beauty (2), fixed (2), representative (2), worlds (2), measured (2), taking (2), procedure (2), pulling (2), hat (2), rabbits (2), almost (2), complexity (2), rigour (2), down (2), empirically (2), deductivist (2), empty (2), wanting (2), austerity (2), utility (2), preferences (2), alarm (2), brain (2), helpful (2), standardly (2), clearing (2), etc (2), motivated (2), validates (2), hear (2), unsettled (2), website, name, email, loading, collapse, bar, manage, subscriptions, site, reader, report, privacy, already, join, 607, subscribers, feeds, entries, 112, archives, 707, varia, 542, 999, 393, politics, 301, 168, categories, list, torssander, fredtorssander, netiquette, anonymous, argumentations, abusive, language, links, please, remember, full, professor, leaves, respond, randomness, representativeness, credible, credibility, revolution, statsskulden, problemet, välförtjänt, bottenrekord, audioholic, summer, stein, paradox, lesson, unbiasedness, love, secret, finding, late, student, stuff, efter, studenten, personal, ekonomiska, politiken, greatest, country, earth, adorno, posts, page, evident, aspects, determining, everywhere, sacrificed, desire, beware, method, crucible, atom, was, obviously, vey, validity, expression, facilitate, clarify, guard, faults, proper, contrast, transient, transformable, scrutinize, latter, attitude, feasible, abandon, quest, replace, positivism, fate, correct, representation, thing, capture, complex, contextual, space, exhibiting, justified, ontological, conditions, possibly, serious, doubts, domain, surface, powers, capacities, persistence, generality, looked, deep, visit, playground, setting, correlations, eventually, mimic, existing, properties, justifiable, entitled, followers, express, features, crucial, ingredient, wants, harm, considering, arguments, probabilites, strictly, pertain, faith, neat, worse, implicit, independence, collide, ugly, validation, adequacy, application, justify, explicitly, define, resources, money, defined, analyzed, convenient, fictions, reliance, widespread, commonly, analyze, convenience, rhetoric, seductive, generated, assumes, handwaving, inadequate, freedman, writes, 2010, 105, 111, observational, instead, creates, parallel, universes, wasn, wonder, ultimately, mathematicians, formal, confront, target, unquestioningly, fisher, von, mises, gibbs, ensemble, series, econometrist, works, treading, shaky, ground, asks, here, bold, considerations, usefulness, deal, conventionally, automatically, ourselves, assign, firm, hicks, neatly, puts, 1979, 120, exactly, start, axiom, postulating, unknown, conceivable, satisfied, autonomous, events, raise, objection, invoke, presupposes, describe, risk, genuine, structures, hidden, truths, discovered, artificial, inventions, consequently, perceived, representations, generating, process, instrumentally, conceived, constructs, realized, searching, degree, autonomy, extrinsic, practically, decisive, relatively, few, proceed, natural, factors, trygve, father, wrote, operations, alone, cycle, consisted, various, cluster, around, shapes, knew, prediction, might, rule, hopefully, persist, somewhat, idiosyncratic, vocabulary, ruled, shall, remain, doubt, outside, change, taught, old, catechism, force, fed, elsewhere, competition, prefer, maximize, interaction, summarized, propensity, organizing, devices, produce, freeing, confining, personally, myself, proud, neoclassicist, minded, approaches, quite, fanatical, defending, maximization, perplexing, grasp, considers, said, emphasis, added, speech, last, gone, astray, contributed, causing, todays, crisis, alternatives, deeply, rightly, certainly, tries, present, himself, radical, anti, establishment, counts, die, teflon, coated, felt, similar, critique, asking, realism, pluralism, blaming, foresee, crises, makers, stick, fine, commenting, cambridge, pontus, interview, heterodox, creationism, medicine, dissed, curriculum, moving, pluralist, direction, having, monumental, challenging, ruling, post, dissing, rethink, andrew, gelman, eric, loken, multiplicities, obvious, highly, unscrupulous, construct, completely, rules, exclusion, coding, thus, needs, conditional, faced, multiple, measures, correctly, produces, noisy, measure, decide, interested, differences, short, expressed, contexts, involving, huge, comparisons, performed, sensitive, single, overarching, interacts, partisanship, affect, skills, choices, decision, taken, mistake, statistic, possibilities, table, event, occurred, differ, iii, articles, equation, output, include, degrees, freedom, numerator, denominator, missing, vanish, forget, performing, importantly, constructions, again, read, probably, confirmed, observed, difference, disconfirmation, independent, reported, count, severe, confirm, problematic, strong, rejected, bias, underlining, equate, entail, judgement, relieve, misspecified, zero, valid, substitutes, narayana, kocherlakota, illustrates, shouldn, blindly, convention, picking, cutoff, second, order, choose, affects, type, costs, kinds, errors, offers, indication, seeing, none, narrow, technical, meaning, discard, above, depend, involved, acting, failing, word, special, thanks, jumping, suppose, looks, exciting, gave, got, grades, control, group, didn, concluding, helped, below, typically, deemed, statistically, borde, statsmakterna, stark, anledning, öka, istället, bedriva, mager, kost, lärarledda, föreläsningar, rekordstora, kategorier, rekryteras, studieovana, hem, svårt, knappare, resursramar, kunna, lösa, dilemmat, krav, meritmässigt, svagpresterande, eftersom, sedan, utbildade, enbart, bråkdel, ingen, djärv, gissning, begåvning, åtminstone, approximativt, normalfördelad, lejonparten, begåvningsmässigt, låg, mittpunkten, normalfördelningskurvan, fem, gånger, samma, knappast, räkna, stor, del, utgörs, individer, ligger, normalfördelningskurvans, mittpunkt, rimligen, torde, innebära, proportionen, svårigheterna, uppnå, traditionellt, högt, ställda, akademiska, kravnivåer, ceteris, paribus, bakgrund, egentligen, anmärkningsvärt, problematiserat, vad, utbildningsexplosionen, leda, lag, rustade, omstruktureringen, skolan, decentralisering, avreglering, målstyrning, tvärtemot, politiska, utfästelser, levererat, eftergymnasiala, utbildningsexpansionen, motsvarande, kunskapskontraktion, hos, stora, ägt, skolpolitik, lett, slår, hårdast, utger, värna, litet, inget, kulturkapital, bagaget, hemifrån, senaste, åren, haft, fullständig, explosion, universitets, sätt, klart, glädjande, doktorander, vårt, utbildningssystem, hade, gymnasister, 1950, talet, utbildningsexpansion, mycket, priset, försämrade, möjligheter, tillgodogöra, högskoleutbildningens, kompetenskrav, fallit, föga, sänkt, upprepade, tillfällen, approcherad, media, apropå, frågor, utöver, usual, suspects, också, försökt, lyfta, problematik, sällan, rädsla, korrekt, lyfts, debatten, bra, rutet, svenska, brottas, två, akuta, hantera, krympande, ekonomi, dåligt, förberedda, göran, arrius, håkan, regnér, linda, simonsen, ifau, institutet, arbetsmarknadspolitisk, utvärdering, utbyggnaden, medfört, studentkvalitet, naturlig, följd, övergången, redan, hög, ungdomar, högsta, betygsintervallen, något, överskott, högpresterande, antas, tillkommer, medför, förkunskaper, kommer, växer, grundskolan, gymnasiets, generellt, lärarna, lägga, tid, hjälpa, utbildningarna, vilket, övrig, undervisning, heller, utesluta, sänks, ökad, genomströmning, dagens, resurssystem, incitament, godkänna, når, dag, tveksamma, syns, universitetskanslersämbetets, kvalitetsutvärderingar, nästan, femte, underkändes, bidragande, orsak, bristande, antalet, ökat, snabbare, tillåter, urholkning, ersättningsbeloppen, räknas, faktiska, kostnadsökningarna, minskas, produktivitetsavdrag, urholkningen, fortsätter, tryck, utbildningsplatser, ensidiga, satsningar, gynnar, varken, samhället, tvivelaktig, nuläget, högskolans, begränsade, budgetutrymme, användas, stärka, utbildningens, bygga, gordon, scolded, sins, commit, tried, partial, redemption, road, salvation, been, seduced, siren, seek, regard, quality, feed, continue, worship, altar, credo, regardless, afraid, competes, lines, macro, colleagues, micro, rests, fly, another, proceeds, impeccable, logic, contradict, historical, record, distinguished, historians, george, stigler, dominant, upon, range, internal, pressures, discipline, subjects, posed, unfolding, developments, goes, add, environment, continues, primarily, curious, relating, presumably, ordering, priorities, brief, physical, material, definitions, emphasize, elements, pushing, definitional, limits, suggest, kornai, sciences, tautologically, deducible, earlier, develop, analytical, tools, improved, observation, measurement, words, hungarian, janos, säger, kris, undrar, posthumanistisk, ansats, belyser, reflekterar, överskrider, varanden, öppnar, mellanrum, dimensioner, subjektivitet, kroppslighet, ömsesidighet, mitt, eget, favoritexempel, vetenskapliga, skojeri, hämtat, författaren, artikeln, mellan, väg, filosofisk, utforskning, mellanrummet, följande, intressanta, programförklaring, forskning, inför, sådant, tyckmyckentrutat, vetenskapligt, blaj, taga, pannan, ekvalist, rytmer, poetiskt, affected, gendered, dwell, shape, materialized, objects, encounter, commuters, visiting, railway, station, masculine, males, potentially, violent, bread, winners, female, mothers, housewives, victim, sexual, violence, objectification, bodies, decent, sexually, available, heterosexual, sammanfattas, engelska, börjar, katten, blir, värre, doktorand, alltså, ägnat, minst, flera, miljoner, skattekronor, besöka, järnvägsstationer, komma, fram, doktorandens, chef, planerat, arbete, chefens, chefer, betygskommitté, externa, granskare, bedömt, skrivit, håller, måttet, avhandlingen, undersöker, järnvägstationer, fysiska, sociala, könsperspektiv, kimstad, pendeltågsstation, norrköpings, järnvägsstation, stockholms, centralstation, ingår, studien, resultaten, påverkar, män, kvinnor, vistas, sprang, nyligen, sammanställning, genusdoktorsavhandlingar, godbitar, särskilt, anmärkningsvärd, nämligen, uppräkningen, handlar, rytm, resande, linköpings, pdf, sammanställningen, sammanfattar, andy, bit, kennel, fetch, sticks, sit, surprised, behaviour, isn, similarly, inaccurately, won, behave, interpreting, parameter, estimates, worrying, confidence, intervals, outcome, linearly, combined, adding, together, optimal, aspires, causes, loads, descriptively, incorrect, well, perils, storytelling, forgets, potential, achieved, attaining, conjectures, experiments, warranted, export, certificate, pretty, worthless, proving, reassuring, believing, fairy, happens, teeth, nobel, prize, winning, fairies, matter, enable, considered, constructing, approximate, thoughts, useful, manner, april, wall, street, james, buchanan, inverse, relationship, quantity, demanded, price, core, embodies, pre, supposition, choice, behavior, sufficiently, allow, physicist, water, uphill, respecting, increases, increase, employment, advanced, equivalent, denial, minimal, consequence, advocates, interests, fortunately, handful, throw, centuries, yet, bevy, camp, following, whores, behold, back, 1992, raised, per, cent, neighbour, state, unchanged, increased, alan, krueger, gathered, fast, food, restaurants, states, turned, decreased, counter, demand, anomalous, backward, sloping, supply, curve, subsequently, stayed, away, cost, friends, instance, job, university, became, angry, publishing, traitors, cause, whole, joseph, stiglitz, prevailed, america, schools, despite, contrary, bears, testimony, unfortunately, policymakers, countries, implement, ideas, fundamentalism, recognises, prophets, modesty, enthralled, rationally, efficient, caused, greedy, unions, government, wages, credit, rationing, adam, smith, efficiency, guided, unseen, forces, ordinary, observers, irrationality, profession, john, maynard, keynes, long, stock, struggling, fundamentals, contest, winner, guesses, judges, sometimes, differently, predict, regularly, systematically, interpreted, competing, utilised, orthodox, misapplication, represent, unanswerable, extremely, unrepresentative, million, chance, selection, source, fiction, propagated, repeated, realizations, imagined, studies, fields, represents, amount, apparent, attempting, artifacts, superpopulations, dubious, highlighted, particularly, questionable, clint, ballinger, gets, feeling, practitioners, automatic, machine, input, picture, scope, certainty, interrelatedness, parts, prevent, constituted, atoms, discretely, distinct, separable, accordingly, deductive, defeating, endeavour, restrict, oneself, prove, closed, evaluate
Text of the page (random words):
ärt att man inte i större utsträckning problematiserat vad utbildningsexplosionen i sig kan leda till eftersom vi för femtio år sedan vid våra universitet utbildade enbart en bråkdel av befolkningen är det ingen djärv gissning under antagande av att begåvning i en population är åtminstone approximativt normalfördelad att lejonparten av dessa studenter begåvningsmässigt låg till höger om mittpunkten på normalfördelningskurvan om vi idag tar in fem gånger så många studenter på våra högskolor och universitet kan vi under samma antagande knappast räkna med att en lika stor del av dessa utgörs av individer som ligger till höger om normalfördelningskurvans mittpunkt rimligen torde detta ceteris paribus innebära att i takt med att proportionen av befolkningen som går vidare till högskola och universitet ökar så ökar svårigheterna för många av dessa att uppnå traditionellt högt ställda akademiska kravnivåer här borde i så fall statsmakterna ha ytterligare en stark anledning till att öka resurserna till högskola och universitet istället för att som idag bedriva utbildningar på mager kost och med få lärarledda föreläsningar i rekordstora studentgrupper med nya kategorier av studenter som i allt större utsträckning rekryteras från studieovana hem är det svårt att se hur vi med knappare resursramar ska kunna lösa dilemmat med högre krav på meritmässigt allt mer svagpresterande studenter statistical significance tests do not validate models 25 oct 2016 at 00 20 posted in economics statistics econometrics 1 comment the word significant has a special place in the world of statistics thanks to a test that researchers use to avoid jumping to conclusions from too little data suppose a researcher has what looks like an exciting result she gave 30 kids a new kind of lunch and they all got better grades than a control group that didn t get the lunch before concluding that the lunch helped she must ask the question if it actually had no effect how likely would i be to get this result if that probability or p value is below a certain threshold typically set at 5 percent the result is deemed statistically significant clearly this statistical significance is not the same as real world significance all it offers is an indication of whether you re seeing an effect where there is none even this narrow technical meaning though depends on where you set the threshold at which you are willing to discard the null hypothesis that is in the above case the possibility that there is no effect i would argue that there s no good reason to always set it at 5 percent rather it should depend on what is being studied and on the risks involved in acting or failing to act on the conclusions this example illustrates three lessons first researchers shouldn t blindly follow convention in picking an appropriate p value cutoff second in order to choose the right p value threshold they need to know how the threshold affects the probability of a type ii error finally they should consider as best they can the costs associated with the two kinds of errors statistics is a powerful tool but like any powerful tool it can t be used the same way in all situations narayana kocherlakota good lessons indeed underlining how important it is not to equate science with statistical calculation all science entail human judgement and using statistical models doesn t relieve us of that necessity working with misspecified models the scientific value of significance testing is actually zero even though you re making valid statistical inferences statistical models and concomitant significance tests are no substitutes for doing science in its standard form a significance test is not the kind of severe test that we are looking for in our search for being able to confirm or disconfirm empirical scientific hypotheses this is problematic for many reasons one being that there is a strong tendency to accept the null hypothesis since they can t be rejected at the standard 5 significance level in their standard form significance tests bias against new hypotheses by making it hard to disconfirm the null hypothesis and as shown over and over again when it is applied people have a tendency to read not disconfirmed as probably confirmed standard scientific methodology tells us that when there is only say a 10 probability that pure sampling error could account for the observed difference between the data and the null hypothesis it would be more reasonable to conclude that we have a case of disconfirmation especially if we perform many independent tests of our hypothesis and they all give about the same 10 result as our reported one i guess most researchers would count the hypothesis as even more disconfirmed we should never forget that the underlying parameters we use when performing significance tests are model constructions our p values mean next to nothing if the model is wrong and most importantly statistical significance tests do not validate models in journal articles a typical regression equation will have an intercept and several explanatory variables the regression output will usually include an f test with p 1 degrees of freedom in the numerator and n p in the denominator the null hypothesis will not be stated the missing null hypothesis is that all the coefficients vanish except the intercept if f is significant that is often thought to validate the model mistake the f test takes the model as given significance only means this if the model is right and the coefficients are 0 it is very unlikely to get such a big f statistic logically there are three possibilities on the table i an unlikely event occurred ii or the model is right and some of the coefficients differ from 0 iii or the model is wrong so why p values cannot be taken at face value 24 oct 2016 at 09 00 posted in economics statistics econometrics 7 comments a researcher is interested in differences between democrats and republicans in how they perform in a short mathematics test when it is expressed in two different contexts either involving health care or the military the research hypothesis is that context matters and one would expect democrats to do better in the health care context and republicans in the military context at this point there is a huge number of possible comparisons that can be performed all consistent with the data for example the pattern could be found with statistical significance among men and not among women explicable under the theory that men are more ideological than women or the pattern could be found among women but not among men explicable under the theory that women are more sensitive to context compared to men a single overarching research hypothesis in this case the idea that issue context interacts with political partisanship to affect mathematical problem solving skills corresponds to many different possible choices of the decision variable at one level these multiplicities are obvious and it would take a highly unscrupulous researcher to perform test after test in a search for statistical significance given a particular data set it is not so difficult to look at the data and construct completely reasonable rules for data exclusion coding and data analysis that can lead to statistical significance thus the researcher needs only perform one test but that test is conditional on the data a researcher when faced with multiple reasonable measures can reason perhaps correctly that the one that produces a significant result is more likely to be the least noisy measure but then decide incorrectly to draw inferences based on that one only andrew gelman eric loken mainstream economists dissing people that want to rethink economics 22 oct 2016 at 20 02 posted in economics 14 comments there s a lot of commenting on the blog now after yours truly put up a post where cambridge economist pontus rendahl in an interview compared heterodox economics to creationism and alternative medicine and totally dissed students that want to see the economics curriculum moving in a more pluralist direction sad to say rendahl is not the only mainstream economist having monumental problems when trying to argue with people challenging the ruling orthodoxy a couple of years ago paul krugman felt a similar urge to defend mainstream neoclassical economics against the critique from students asking for more relevance realism and pluralism in the teaching of economics according to krugman the students and people like yours truly are wrong in blaming mainstream economics for not being relevant and not being able to foresee crises to krugman there is nothing wrong with standard theory and economics textbooks if only policy makers and economists stick to standard economic analysis everything would be just fine i ll be dipped if there s anything the last couple of years have shown us it is that economists have gone astray krugman s standard theory mainstream neoclassical economics has contributed to causing todays s economic crisis rather than to solving it reading krugman i guess a lot of the young economics students that today are looking for alternatives to mainstream neoclassical theory are deeply disappointed rightly so but although krugman especially on his blog certainly tries to present himself as a kind of radical and anti establishment economics guy when it really counts he shows what he is a die hard teflon coated mainstream neoclassical economist perhaps this becomes less perplexing to grasp when one considers what krugman said in a speech emphasis added in 1996 i like to think that i am more open minded about alternative approaches to economics than most but i am basically a maximization and equilibrium kind of guy indeed i am quite fanatical about defending the relevance of standard economic models in many situations personally i consider myself a proud neoclassicist by this i clearly don t mean that i believe in perfect competition all the way what i mean is that i prefer when i can to make sense of the world using models in which individuals maximize and the interaction of these individuals can be summarized by some concept of equilibrium i have seen the propensity of those who try to do economics without those organizing devices to produce sheer nonsense when they imagine they are freeing themselves from some confining orthodoxy so now all young economics students that want to see a real change in economics and the way it s taught now you know where you have people like rendahl and krugman if you really want something other than the same old mainstream neoclassical catechism if you really don t want to be force fed with mainstream neoclassical theories and models you have to look elsewhere share on facebook opens in new window facebook share on x opens in new window x share on bluesky opens in new window bluesky econometrics and the bridge between model and reality 21 oct 2016 at 23 28 posted in statistics econometrics 1 comment trygve haavelmo the father of modern probabilistic econometrics wrote that he and other econometricians could not build a complete bridge between our models and reality by logical operations alone but finally had to make a non logical jump statistical testing of business cycle theories 1943 15 a part of that jump consisted in that econometricians like to believe that the various a priori possible sequences would somehow cluster around some typical time shapes which if we knew them could be used for prediction 1943 16 but since we do not know the true distribution one has to look for the mechanisms processes that might rule the data and that hopefully persist so that predictions may be made of possible hypothesis on different time sequences samples in haavelmo s somewhat idiosyncratic vocabulary most had to be ruled out a priori by economic theory although one shall always remain in doubt as to the possibility of some outside hypothesis being the true one 1943 18 the explanations we can give of economic relations and structures based on econometric models are according to haavelmo not hidden truths to be discovered but rather our own artificial inventions models are consequently perceived not as true representations of the data generating process but rather instrumentally conceived as if constructs their intrinsic closure is realized by searching for parameters showing a great degree of invariance or relative autonomy and the extrinsic closure by hoping that the practically decisive explanatory variables are relatively few so that one may proceed as if natural limitations of the number of relevant factors exist the probability approach in econometrics 1944 29 but why the logically conceivable really should turn out to be the case is difficult to see at least if we are not satisfied by sheer hope in real economies it is unlikely that we find many autonomous relations and events and one could of course also raise the objection that to invoke a probabilistic approach to econometrics presupposes e g that we have to be able to describe the world in terms of risk rather than genuine uncertainty and that is exactly what haavelmo 1944 48 does to make this a rational problem of statistical inference we have to start out by an axiom postulating that every set of observable variables has associated with it one particular true but unknown probability law but to use this trick of our own and just assign a certain probability law to a system of observable variables however cannot just as little as hoping build a firm bridge between model and reality treating phenomena as if they essentially were stochastic processes is not the same as showing that they essentially are stochastic processes as hicks so neatly puts it in causality in economics 1979 120 21 things become more difficult when we turn to time series the econometrist who works in that field may claim that he is not treading on very shaky ground but if one asks him what he is really doing he will not find it easy even here to give a convincing answer h e must be treating the observations known to him as a sample of a larger population but what population i am bold enough to conclude from these considerations that the usefulness of statistical or stochastic methods in economics is a good deal less than is now conventionally supposed we have no business to turn to them automatically we should always ask ourselves before we apply them whether they are appropriate to the problem in hand and as if this wasn t enough one could also seriously wonder what kind of populations these statistical and econometric models ultimately are based on why should we as social scientists and not as pure mathematicians working with formal axiomatic systems without the urge to confront our models with real target systems unquestioningly accept haavelmo s infinite population fisher s hypothetical infinite population von mises s collective or gibbs s ensemble of course one could treat our observational or experimental data as random samples from real populations i have no proble...
|
