Keep it sophisticatedly simple
Open accessAbstract
Introduction Some years ago, I came upon the phrase used in industry, ‘Keep it simple stupid’, that is, KISS, and thought about it in relation to scientific model-building. Since some simple models are stupid, I decided to reinterpret KISS to mean ‘Keep it sophisticatedly simple.’ In any event, KISS is very popular in many scientific and non-scientific areas. For example, the slogan of the Honda Motor Company is, ‘We make it simple.’ The Dutch Schipol airport in its advertising claims that, ‘It excels because it is simple and convenient.’ And it is well known that Einstein advised in connection with theorizing in the natural sciences, ‘Make it as simple as possible but no simpler.’ Also, the famous physicist Jaynes (1985, p. 344) wrote, ‘We keep our model as simple as possible so as not to obscure the point to be made and also to heed Arnold Zellner's wise advice about “sophisticatedly simple” models.’ Many, including myself, have for long advocated that workers in econometrics and statistics follow the advice of natural scientists and others to keep analyses and models sophisticatedly simple. In addition, I have pointed out that there are many important, sophisticatedly simple models and methods that work well in practice, that is in explanation and prediction, namely s = ½ gt 2 , E = mc 2 , PV = RT, maxent, etc.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The Federal Reserve Bank of Minneapolis scrapped its complicated vector autoregressive models after their poor performance in forecasting turning points.
“he reported 'the Federal Reserve Bank of Minneapolis decided to scrap its complicated vector autoregressive (V AR, i.e., Very Awful Regression) models after their poor performance in forecasting turning points, etc.'”
From Illusions in Regression Analysis (2011) · cites this paper as Zellner (2001)
The Abstract says the Federal Reserve Bank of Minneapolis was revising its VAR models after poor turning-point forecasts, not scrapping them. It separately says Federal Reserve officials decided to scrap the Federal Reserve–MIT–PENN model; the claim conflates those two accounts. (LLM, conf=0.90)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredTypically, fit improves as complexity increases, while ex ante forecast accuracy decreases.
“Typically, fit improves as complexity increases, while ex ante forecast accuracy decreases - a conclusion that Zellner (2001) traced back to Sir Harold Jeffreys in the 1930s.”
From Illusions in Regression Analysis (2011) · cites this paper as Zellner (2001)
Section IV reports that reducing model complexity improved out-of-sample forecasts, and the Introduction describes Jeffreys’s preference for simpler models. But our copy does not say that fit typically improves as complexity increases or that Jeffreys identified the claimed fit-versus-forecast relationship. Section V also notes that well-chosen added complexity can improve prediction. (LLM, conf=0.78)
Experimental — accuracy not yet measuredZellner reached the same conclusion that increased complexity of regression models typically reduced forecast accuracy.
“Zellner (2001) reached the same conclusion in his review of the research.”
From Illusions in Regression Analysis (2011) · cites this paper as Zellner (2001)
The Abstract reports that complicated macroeconometric models often forecast unsatisfactorily and, in some cases, no better or worse than simple time-series models. That supports Zellner’s preference for simplicity, but not the broader claim that increasing the complexity of regression models typically reduces forecast accuracy. The Summary and Conclusions also says well-chosen added complexity may improve prediction. (LLM, conf=0.84)
Experimental — accuracy not yet measuredFit bears little relationship to ex ante forecast accuracy, especially for time series.
“This ignores the research showing that fit bears little relationship to ex ante forecast accuracy, especially for time series.”
From Illusions in Regression Analysis (2011) · cites this paper as Zellner (2001)
The Abstract (page unavailable) reports that complicated macroeconometric models often forecast no better than simple time-series models. Section IV (page unavailable) also evaluates out-of-sample forecasts. Neither passage establishes that model fit bears little relationship to ex ante forecast accuracy, so the claim overstates what this source shows. (LLM, conf=0.77)
Experimental — accuracy not yet measuredThe index model allows analysts to take account of 'the knowledge present in a field' as recommended by Zellner.
“The index model allows analysts to take account of 'the knowledge present in a field' as recommended by Zellner (2001).”
From Illusions in Regression Analysis (2011) · cites this paper as Zellner (2001)
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 18
- R.L. Coope r, 'The Predic tive Perfor mance of Quarte rly Econom etric Model s of the U.S. Econo my,' in B.G. Hickm an (ed.), Econom etric Model s of Cyclic al Behavi or, Vol. 2 (New York: Colum bia U. Press, 1972), 813-9 26: Fried man, 'Comm ent on 'A Test of an Econom etric Model for the United States, 1921- 47' by Carl Christ, ' in Confer ence on Busine ss Cycles (New York: Nation al Burea u of Econo mic Resear ch, Inc., 1951), 107-11
- M. Fried man, A Theor y of the Consu mptio n Functi on (Princ eton: Prince ton U. Press, 1957)
- A. Garcia -Ferre r, R.A. Highfi eld, F.C. Palm and A. Zenne r, 'Macr oecon omic Forcas ting Using Poole d Intern ationa l Data,' Journa l of Busine ss and Econo mic Statist ics 5 (1987) , 53-67.
- B.G. Hickm an (ed.), Econom etric Model s of Cyclic al Behavi or (New York) Colum bia U. Press, 1972)
- C. Hong, 'Forec asting Real Output Growt h Rates and Cyclic al Proper ties of Model s: A Bayesi an Approa ch,' Ph.D. thesis, Dept. of Econo mics, U. of Chica go (1989) .
- H. Jeffreys, Theory of Probability (Oxford: Oxford U. Press, 1939, 1948, 1961,1967, 1988).
- Jeffreys, Scientific Inference (Cambridg e: Cambridge U. Press, 1931, 1957).
- H.A. Keuzenkam p and M. McAleer, 'Simplicity , Scientific Inference and Econometri c Modelling, ' Economic Journal 105 (1995), 1-21. .A. Keuzenkam p and M. McAleer, 'The Complexity of Simplicity, " Mathematic s and Computers in Simulation , 43 (1997), 553-561.
- R.B. Litterman, 'Forecastin g with Bayesian Vector Autoregress ions: Five Years of Experience ,' J. of Business and Economic Statistics, 4 (1986), 25-38 (with discussion) .
- R. Meese and K. Rogoff, 'The Out of Sample Failure of Empirical Exchange Rate Models: Sampling Error or Misspecific ation? In J. A. Frenkel (ed.) Exchange Rates and Internation al Economics , (Chicago: U. of Chicago Press, 1983).
- A.Zellner and M.S. Geisel, 'Analysis o ill istributed Lag Models with Applications to Consumption Function Estimation,' Econometrica 38 (1970), 865-888.
- A. Zellner and F.Palm, 'Time Series Analysis and Simultaneous Equation Econometric Models,' J.of Econometrics 2 (1974), 17-54.
- A. Zellner and C. Hong, 'Forecasting International Growth Rates Using layesian Shrinkage and Other Procedures," J. of Econometrics, Annals, P. Schmidt (ed.) Issues in Econometric Forecasting, 40 (1989), 183-202.
- A. Zellner and C. Hong, `Bayesian Methods for Forecasting Turning Points in Economic Time Series: Sensitivity of Forecasts to Asymmetry of Loss Structures,' in K. Lahiri and G. Moore (ed.), Leading Indicators: New Approaches and Forecasting Records (Cambridge: Cambridge U. Press, 1991), 129-140.
- Zellner and S. Peck, 'Simulation Experiments with a Quarterly Macroeconometric Model of the U.S. Economy,' in A.A. Powell and R.A. Williams (eds.), Econometric Studies of Macro and Monetary Relations, (Amsterdam: North-Holland, 1973), 149-168. .
- A. Zellner C. Hong and G. M. Gulati, 'Turning Points in Economic Time Series, Loss Structures and Bayesian Forecasting," in S. Geisser, J. Hodges, S.J. Press and A. Zellner (eds.), Bayesian and Likelihood Methods in Statistics and Econometrics: Essays in Honor of George A. Barnard (Amsterdam: North-Hollan d), 371-393.
- Zellner, C. Hong and C. Min, 'Forecasting Turning Points in International Output Growth Rates Using Bayesian Exponentially Weighted Autoregression , Time-Varying Parameter, and Pooling Techniques,' J. of Econometrics, 49 (1991), 275-304.
- Zellner and H. Ryu, 'Alternative Functional Forms for Production, Cost and Returns to Scale Functions,' J.of Applied Econometrics, 13 (1998), 101-127.
Cited by
Sorted by citations
Illusions in Regression Analysis
Armstrong, J. S. (2011) · doi:10.2139/ssrn.1969740 · Cited by 73 · see citing works
- Authors
- Zellner, A.
- Affiliations
- University of Chicago (US)
- Corresponding author
- Zellner, Arnold
- Venue
- Simplicity, Inference and Modelling
- Publisher
- Cambridge University Press
- Published
- 2002-02-07
- Pages
- 242-262
- OpenAlex
- W2501338012
- Cited by
- 59
- References
- 0
- Citations by year
- 2025: 3 · 2024: 3 · 2023: 1 · 2022: 3 · 2021: 7 · 2020: 2 · 2019: 5 · 2018: 2 · 2017: 3 · 2015: 4 · 2014: 2 · 2013: 1 · 2012: 4
- FWCI
- 9.69
- Type
- book chapter
- Language
- en
- Open access
- green
- ageconsearch.umn.edu
- Licence
- Locations
- Cambridge University Press eBooks (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- AgEcon Search (University of Minnesota, USA) (submittedVersion, cc-by-nc-nd)
- AgEcon Search (University of Minnesota, USA)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Philosophy and History of Science
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Keywords
- model simplicity, econometric model selection, statistical model building, parsimony, maximum entropy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
