Samenvatting
Economic methodology: Summary of the lectures (GRADE: 8.9)
summary of the lectures of economic methodology, year 2 UvA
[Meer zien]Voorbeeld 4 van de 45 pagina's
Voorbeeld 4 van de 45 pagina's
In winkelwagengesponsord bericht van onze partner
Voorbeeld van de inhoud
2Summary of the lecturesLecture 1
Science is the result of scientific research (explanations, descriptions, measurements, predictions,
recommendations)
Science is scientific if it meets certain criteria.
Received view: science better claim to knowledge because accepts only facts.
Falsificationism: as long as we fail, theory is right
Ontology: field of science, philosophy that deals w/reality of the world. Ex; is there a god, what is
reality like
Epistemology: about theory of knowledge. What criteria does knowledge have to fulfill in order to
constitute knowledge
Method: if you have evidence on business studies, you have business method. Use when talk about
specific techniques to gather knowledge (not methodology)
Sociology of sciences: studying what scientists actually do. Describe for ex, what ontology is in specific
field, just describing not judging
Positive science (facts): describe, explain, predict
Normative science (norms): prescriptions/recommendations
Stick to facts, if you say something that isn’t checkable, you’re not a scientist
Science must be meaningful. Meaningful is empirically verifiable (synthetic a posteriori, true by
observation, scientific, meaningful) and/or logically verifiable (analytic, true by definition, scientific,
meaningful). Not meaningful is synthetic a priori (so not scientific, not meaningful)
Two foundations of logical empiricism:
If LP confronted with scientific theory, need to take 2 steps:
1. make logical analysis: reconstruct argument in formal logical mathematic terms, you can find if
there is inconsistency in theory
2. empiricism: make sure all terms are observable in theory (measurable) and then do
observation/measurement, and then check if whole story is true.
If can say yes to both, scientific theory, if not to both: not scientific. Implication: large parts of psycho
analysis/history writing not scientific. Logical positivist very radical. Forget psychology, languages,
history because all of them not sufficiently observable & measurable.
Operationalization: terms/theoretical category linking up with observational category called
operationalization. Difficult to see what for, example: investment. Define investment in such a way =
operationalization that is observable, so that if we talk about investment, we all mean the same thing.
All theoretical categories need to be operationalized. Another example: define clock, sunrise 24
hours, so make clock that ticks away 24 hours, then use clock to see measurement of time.
Correspondence rule??
When are you looking for work? Its vague, needs to be operationalized in terms of unemployed
labour force.
Scientific statements are either synthetic a posteriori or analytic
Couple examples scientific/non-scientific statements.
The earth revolves around the sun: scientific, synthetic a posteriori. Earth and sun assumed to be
sufficiently operationalized so we all know what we talk about. We have seen this by using telescope
etc (true by observation)
The subconscious is the origin of the world: not scientific, synthetic a priori, not true by observation
and not true by definition
If the price of a good rises, demand for that good falls: scientific, synthetic a posteriori, if price and
demand have been sufficiently operationalized (needs to be assumed in these exercises). If agreed
how we measure them, its synthetic a posteriori, its verifiable in principle. Means something about
the world.
If producer confidence rises, the IS curve shifts outwards: scientific, analytic, its not about empirical
reality, we say if certain something happends, perhaps we can measure produced confidence, then IS
curve shifts outwards. Doesn’t mean anything about world, purely analytic. When link model back to
world, becomes synthetic a posteriori. If model works in certain way, it’s a definition.
, Involuntary unemployment does not exist: not scientific, synthetic a priori, NOT can never prove that
is does not exist. Perhaps someone somewhere is employed w/o choosing to be. On basis of
observation can never claim something does not exist. Can never research all potential instances.
Demarcation: distinguish between scientific and non-scientific (science and nonscience). Everything is
related: can never observe everything and relate it to anything else, cannot operationalize.
Assessment: (context of discovery and context of justification). Science only needs to be assessed,
judge scientific or not after it has been written down. not relevant how you get to your theory
(discovery). Acc to LP, not relevant for judgement of theory & not interesting for scientists.
Whats interesting: theory that has been written down and can be checked by empirical observation
(justification), so only justification matters, only that one is checked
Context of discovery: how are you going to do your research
Context of justification: what have you written down, results. Need to be checkable by observation.
Scientific if:
1. logically consistent
2. all terms operationalized (so e.g experiment can be set up to verify statement).
Deductive nomological explanation (DN-model):
To structure what scientific explanation should look like. Starts with phenomenon to explain (law of
Archimedes), then law-like propostion: perhaps law of Archimedes has something to do with that of
that form,
If see price of coffee rising, then we have law-like propostion that wherever supply falls and demand
stays equal, price rises = law of demand. Need to figure out if supply is indeed lower can coffee. Due
to climate changes, supply of coffee more costly.
Can also predict things: if start with law-like proposition law of demand, and see supply of coffee is
falling, can predict price is rising,
Symmetry thesis:
If we start w/1 and 2, we predict 3
If we start with 3, and then fill in 1 and 2, we’re explaining 3
Lawlike proposition:
Until now: Something only scientific if its truth can be verified by observation
Problem: now they say, to explain things we need a law. If we need to explain why price of coffee has
risen, we already assume this law holds for coffee as well, we haven’t observed this because
otherwise law wouldn’t explain why coffee is rising in price. So problem with law-like propostion is
that you claim its true, for things that by definition you haven’t observed yet. If you’re predicting
this is obvious, using law of demand to predict price rise and you haven’t observed price rise yet, so
predicting price rise is not scientific.
They want to get rid of statements that are not directly checkable by observation, yet we want
science to explain, but to explain things and predict things they need laws. In order for law to
explain/predict things law needs to be applicable to phenomenon that hasn’t happened yet in case of
predicting or hasn’t been explained yet. In both cases cannot just assume law holds (not on basis of
observation)
, Verification of laws: Human problem of induction (black swans)
Conclusion: humean problem of induction, scientific laws are not scientific. Something can only be
scientific if you can ascertain its truth by observation laws never scientific (law: all swans are white).
What you thought was general law is not true at all because of the black swan. Big problem:
1. Instrumentalism: universal statement = instrumental guideline ≠ meaningful, meaningful statement
= certain statement.
Can me instrumental guidelines, but cant claim they are scientific. Meaningful statements are only the
certain ones.
2. confirmationism: proability instrad of certainty, probabilistic explanation. Laws had to be
exceptionless, abandoned.
Lecture 2
Lecture 1 was about logical positivism: scientific knowledge has to be (demarcation criterion):
1. logically consistent (analytical statements)
2. verifiable by observation when empirical (requires operationalization of theoretical terms by
correspondence rules): synthetic a posteriori statements.
- explanation and prediction DN model.
Truth should only depend on observation. End product should be verifiable, only then its science.
Econometrics:
1. first, model must be specified in explicit functional – often linear – form. Mathematical model to
ascertain. Logical consistency
2. second step is to decide on appropriate data definitions, and to assemble the relevant data series
for those variables included in the model. Use statistics, its help with operationalization. Make sure all
terms are verifiable
3. third step is to form a bridge between theory and data through use of statistical methods. Bridge
consists of various sets of statistics, which help to determine the validity of the theoretical model.
Establish validity of the model.
Positivism in econometrics:
- mathematics as analytical foundation (model structure) step 1
- statistics as observational foundation (model test): operationalization and data choice guarantee
verifiability (step 2), statistical testing verifies (step 3)
- explanation and prediction follows DN-model
Keynes-tinbergen debate
Point Keynes is making: what you’re doing is not testing hypotheses, you’re measuring correlations.
You’re not testing whether certain correlation is true, you measure correlation on basis of your
research, it isn’t testing. If measuring rather than testing, measurements will only be correct if start
with complete model:
, Example: if Y=C+I, if forget G, relative importance of government is put into parameters of c and I, so
parameters overestimated influence of remaining variables. If factors affects negatively and kept out
of model, you underestimate them (because affects parameters) vice versa. If not taken into account
always over or underestimating.
Keynes’ criticism on tinbergen’s econometric model
1. Tinbergen does not test but measures
2. Tinbergen’s measuring is only possible is all causalities are known (complete list/model).
Measurements will never be correct if model is not complete
(Haavelmo) Probability approach in econometrics/laws in economics
“whether or not we might hope to find elements of invariance in economic life, upon which to
establish permament “laws””(Haavelmo,1944)
If we can find model complete enough to stabilize parameters we find ourselves economic laws.
Make as many measurements as needed, until at some point parameters we are measuring are no
longer variant (no longer behave eratically) -> find yourself a law
Take variables that are all important, establish linear forms, try and find situations in which lot of
variables happen to be stable
Cannot test but only measure, to establish measurements that are stable you need to find economic
laws. If at some point we find certain variable that does not seem to influence economy at all, might
believe variable not significant/relevant, should be taken out of model. It does matter if reach certain
threshold. If = 0, variable not relevant, parameter = 0 or variable might be positive if variable you’re
interested in reaches certain threshold. No way to tell whether to keep in variable of leave out. that’s
the problem of passive observation (=if don’t see variable influence dependent variable may mean
variable hasn’t changed enough but still has potential influence, or it may mean variable is not
important at all)
Haavelmo: goal is measuring, but of and by means of probabilities, not exact values.
- you measure probabilities and not exact values. But we never know if certain variables we do not
take into account because we don’t see any influence we see correlation of 0, will never have any
influence or didn’t have any influence at that time.
- what can we do? Wait to become important and then include in model? Always repairing model
after the fact which isn’t good, or start out with as comprehensive model as possible ( So put
everything in model of fear of missing out).
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper kdamsterdam. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €5,99. Je zit daarna nergens aan vast.
Is Stuvia te vertrouwen?
4,6 sterren op Google & Trustpilot (+1000 reviews)
Afgelopen 30 dagen zijn er 79202 samenvattingen verkocht
Opgericht in 2010, al 14 jaar dé plek om samenvattingen te kopen
