Page 1 of 5
Advances in Social Sciences Research Journal – Vol. 9, No. 3
Publication Date: March 25, 2022
DOI:10.14738/assrj.93.11994. Bechtel, G. G. (2022). Global Assets Mitigate Global Inflation. Advances in Social Sciences Research Journal, 9(3). 154-158.
Services for Science and Education – United Kingdom
Global Assets Mitigate Global Inflation
Gordon G. Bechtel
Warrington College of Business, University of Florida Gainesville, USA
ABSTRACT
This article finds that 26 global assets mitigate global inflation. Fractional
polynomial regressions return sizable goodness-of-fit R2s for these assets, along
with high mitigating Pearson correlations between these assets and global inflation.
It is hoped that future fractional polynomial regressions will reveal other global
assets that also mitigate global inflation. In view of Russia's invasion of the Ukraine,
which preempted American President Biden's State of the Union address
(Aljazeera, March 1 2022), and the trade war between the United States and China,
the United Nations has now stepped in to protect it's member states from the
interest-rate and inflationarity effects wrought by these three super powers.
Keywords: Computing regression coefficients and their powers; Global population
weighting; United-Nations, World-Bank, and Swiss-Economic-Institute time series.
INTRODUCTION
In February , 2022 UN Secretary General Antioio Guterrus railed "we cannot even foresee in
relation to the consequences for the global economy in a moment when we are emerging from
the COVID (pandemic) and so many developing countries absolutely need to have space for the
recovery which would be very, very difficult, with the high prices of oil, with the end of exports
of wheat from Ukraine, and with the rising interest rates caused by instability in international
markets”.
Accordingly, the first purpose of the present paper is to situate rising interest rates and global
inflation in the new data science. Our variables are measured on ratio and identity scales, which
exceed interval scales in the hierarchy of scientific measurement. For almost half a century
interval scales have been beset with skepticism about their incremental benefits over and
above ratio and identity scales already in use. The problems associated with interval scaling,
i.e. survey sampling, questionnaire interrogation, probabilistic inference, and significance
testing, are absent with the global assets measured here in dollars, per cents, counts, cubic
meters, and volume (cf. Section 6). In addition, the host of long-standing, and now acute, issues
daunting micro-data collection and analysis are evaded by the fractional polynomial
regressions in Stata commands (1) and (2) below. Second, we introduce the concept of
fractional polynomial causation and contrast it with Clive Granger’s cointegration, detailed in
his 2023 Nobel Lecture, between the time series of an important dependent variable and that
of an independent variable thought to predict it. This contrast points up the significance of the
major finding of this paper: namely; 26 global assets mitagate global inflation.
Page 2 of 5
155
Bechtel, G. G. (2022). Global Assets Mitigate Global Inflation. Advances in Social Sciences Research Journal, 9(3). 154-158.
URL: http://dx.doi.org/10.14738/assrj.93.11994
GLOBAL INFLATION IN THE NEW DATA SCIENCE
The new data science, as well as venerable global media, rest upon crucial time series such as
that of global inflation. Our analysis here of a global inflation time series over 1991-2017
overrides the basic canon of statistical inference, that there is fundamental uncertainty in all
data. Neither denying nor quantifying uncertainty, we simply ignore it This approach to
sequential time series brings compelling advantages to the new data science. Probabilistic
inference is replaced by parameter computation and random variables give way to real
variables. This suggests further “statistical thinking and new foundational frameworks” that
help sort out “the many philosophical issues data science presents ... “ [Davidian, 2013]. This
call has been echoed by the American National Science Foundation, who has “released a revised
version of the solicitation ‘Critical Techniques and Technologies for Advancing Foundations
and Applications of Big Data Science ... ‘ ” (Vogelius et al., 2015).
The variables chosen here to affect global inflation are measured on ratio and identity scales,
which exceed interval scales in the hierarchy of scientific measurement (Stevens, S. S.; 1946,
Torgerson, W.S. 1958, Suppes, P.; Zinnes, J. L., 1963). For almost half a century interval scales
have been beset with skepticism about their incremental benefits over and above ratio and
identity scales already in use (Shapiro, 1972) (p. 373). The problems associated with interval
scaling, i.e. survey sampling, questionnaire interrogation, probabilistic inference, and
significance testing, are absent from the global assets measured here in dollars, per cents,
counts, cubic meters, and volume. In addition, the host of long-standing, and now acute, issues
daunting micro-data collection and analysis are evaded by the fractional polynomial regression
in Stata command (1) below. The results brought by this regression advance the mitigating
effects of global assets on global inflation.
THEORY: FRACTIONAL POLYNOMIAL CAUSATION
Research into cause and effect in the last century was followed by Clive Granger’s 2003 Nobel
Lecture, which treated nonlinearity between a time series X and a time series Y. Relationships
between X and Y are exhibited and discussed daily on worldwide television (cf. Aljazeera, CNBC,
Shanghai Media Group, and Taipei Times). Clive Granger’s timely Nobel Lecture introduced the
concept of co-integration between X and Y. Co-integration between time series X and time
series Y exists when the patterns in X are approximately repeated in Y after some time lag.
Thus, past values of X can be used for the prediction of future values of Y
(https://en.wikipedia.org/wiki/Granger_causality). We leave it to the reader to decide if co- integration between Y and X, or 2-way-FP regression of Y on X and X on Y, is more precise, as
well as more originative of new questions. We also note that a search of recent literature
reveals no mention of fractional polynomial nonlinearity as a specific departure from
nonlinearity (cf. Barnett et. al, 2000 and Kumar, 2019). This specificity, along with FP causality,
are developed below. Replacing X by and Y by , we have evaluate R2s by their nearness to one.
criterion of FP causation is especially stringent because it is restricted to just one independent
variable. The measurement of and is provided by the following definitions:
Table 1 lists the United-Nation’s, World-Bank’s, and Swiss-Economic-Institute’s measurements
of 26 global assets studied in the present article.
Page 3 of 5
156
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 3, March-2022
Services for Science and Education – United Kingdom
Table 1. Global Inflation and Global Assets
-----------------------------------------------------------------------------------------
γ : Global GDP (trillions of currentUS$) κ:
KOF Globalization Index (%)
ι: Global Inflation (%) A:
Human Development Index (%) B:
Stocks traded, total value (% of GDP)
C: Air transport, registered carrier departures worldwide (count)
D: Fixed telephone subscriptions per 100 people (count)
E : Mobil cellular subscriptions per 100 people (count)
F : Renewable internal freshwater resources per capita (cubic meters) G
: Water per capata (volume) H
: Non-Poverty headcount ratio at $1.90 a day 2011 PPP (% of population) I
: Market capitalization of listed domestic companies (trillions of current US$)
J : Merchandise exports (currentUS$)
K : Merchandise imports (currentUS$)
L: Merchandise trade (% of GDP) M:
Ores and metals exports (% of Merchandise exports)
N: Non-military expenditure (% of GDP)
O : Patent applications, nonresidents (count)
P : Patent applications, residents (count)
Q : Trademark applications, direct nonresident (count)
R : Trademark applications, direct resident (count)
S : Exports of goods and services (% of GDP)
T : Imports of goods and services (% of GDP) U :
Global employment (% of total labor force)
V : Global life expectancy (years) W
: Global investment in non-financial assets (% of GDP)
X : Global arable land (% of global land area)
-----------------------------------------------------------------------------------------
METHOD
The fractional polynomial causation of ι by I, is confirmed by R2 = .6162 when time series ι is
FP regressed on time series I. This result is delivered by the following FP regression in Stata
syntax:
fracpoly reg ι I [iweight = GlobalPopulation], degree(1) noscaling (1) The
qualifier [iweight = GlobalPopulation] is an importance weight assigned to the global
population for ι’s and I’s yearly values in 1991 ... 2017. The noscaling option keeps ι scaled in
per cent and I scaled in trillions of current US$. The degree (1) option allows ι to be affected by
only one covariate I. These options for fractional polynomial regression are detailed in
StataCorp. (2011).
RESULTS
Table 2 shows the mitigating R2s delivered by fractional polynomial regressions of ι on the 26
global assets γ κ A B ...T U V W X (cf. Royston and Altman, 1994). These global assets are
Page 4 of 5
157
Bechtel, G. G. (2022). Global Assets Mitigate Global Inflation. Advances in Social Sciences Research Journal, 9(3). 154-158.
URL: http://dx.doi.org/10.14738/assrj.93.11994
measured in dollars, per cents, counts, cubic meters, and volume on a 27 (years) x 26 (global
assets) spreadsheet (cf. Table 1).
Table 2. Regression R2s for ι on Global Assets γ κ A B ...T U V W X
----------------------------------------------------------------------------------------------
.4354 .4608 .4282 .5495 .5505 .3090 .5604 .5499 .5499 .5169 .6162 .4676 .4808
.3913 .0240 .3552 .5789 .5643 .4021 .5028 .4557 .4305 .3555 .4995 .2883 .1169
-----------------------------------------------------------------------------------------------
Stata command (1) also produces the function
ι = f(I) = 35.64485*I-.5 - 1.7254191 (2)
Function (2) shows that the FP regression of global asset I on ι over 1991 ... 2017. Function (2)
of ι on I illustrates the other 25 functions, each of which have different forms. It is important to
note that the R2s in Table 2 are invariant with respect to the units in which global assets γ κ A
B ...T U V W X are calibrated.
Table 3. Product-moment correlations between ι and γ κ A B ...T U V W X
-------------------------------------------------------------------------------------------------------------
-.5662 -.6641 -.6685 -.7105 -.7118 -.4880 -.6113 .7352 .7352 .6950 -.7308 -.5364 -.5397
-.5310 .1236 -.6064 -.7140 -.6523 -.6371 -5694 -.6246 -.6003 .6217 -.7050 -.5444 -.2931
-----------------------------------------------------------------------------------------------------------
SUMMARY
This article finds that 26 global assets mitigate global inflation. Fractional polynomial
regressions return sizable goodness-of-fit R2s for these assets, along with high mitigating
Pearson correlations between these assets and global inflation. It is hoped that future fractional
polynomial regressions will reveal other global assets that also mitigate global inflation. In view
of Russia's invasion of the Ukraine, which preempted American President Biden's State of the
Union address (Aljazeera, March 1 2022), and the trade war between the United States and
China, the United Nations has now stepped in to protect it's member states from the interest- rate and inflationarity effects wrought by these three super powers. In February 2022 UN
Secretary General Antioio Guterrus said "we cannot even foresee in relation to the
consequences for the global economy in a moment when we are emerging from the COVID
(pandemic) and so many developing countries absolutely need to have space for the recovery
which would be very, very difficult, with the high prices of oil, with the end of exports of wheat
from Ukraine, and with the rising interest rates caused by instability in international markets”.
The first purpose of the present paper is to situate rising interest rates and global inflation in
the new data science. Our variables are measured on ratio and identity scales, which exceed
interval scales in the hierarchy of scientific measurement. For almost half a century interval
scales have been beset with skepticism about their incremental benefits over and above ratio
and identity scales already in use. The problems associated with interval scaling, i.e. survey
sampling, questionnaire interrogation, probabilistic inference, and significance testing, are
absent with the global assets measured here in dollars, per cents, counts, cubic meters, and
volume (cf. Table 1). In addition, the host of long-standing, and now acute, issues daunting
micro-data collection and analysis are evaded by the fractional polynomial regressions resulted
here. Second, we introduce the concept of fractional polynomial causation and contrast it with
Clive Granger’s cointegration, detailed in his 2023 Nobel Lecture, between the time series of an
important dependent variable and that of an independent variable thought to predict it. This
Page 5 of 5
158
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 3, March-2022
Services for Science and Education – United Kingdom
contrast points up the significance of the major finding of this paper: namely; 26 global assets
mitagate global inflation.
ACKNOWLEDGMENTS
This article is dedicated to the memory of the author’s best critic, Maria Cohn Bechtel. The
author thanks Timothy Bechtel for contributing the milieu for globalization and GDP, Dr.
Bethany Bechtel for her insistence on monitoring globalization and GDP over time, and Aryil
Bechtel for his suggestion that GDP can be parsimoniously predicted by the KOF Index only.
The author appreciates the reviewers suggestions, which have improved this work.
References
(1) Davidian, M. 2013. Aren’t we data science? Amstat News, July, 3-5.
(2) Vogelius, M.; Kannan, N.; Huo, X. 2015. NSF big data funding opportunity for the statistics community. Amstat
News, April, 6.
(3) Stevens, S. S. 1946. On the theory of scales of measurement. Science, 677-680.(8) Torgerson, W. S. 1958.
Theory and Methods of Scaling; John Wiley & Sons, New York, NY, USA.
(4) Suppes, P.; Zinnes, J. L. 1963. Basic measurement theory; In Handbook of Mathematical Psychology, Volume I;
Bush, R. R., Luce, R. D., Galanter, E., Eds.; John Wiley & Sons, New York, NY, USA, 1-76.
(5) Shapiro, H. 1972. The index of consumer sentiment and economic forecasting: A reappraisal; In Human
Behavior in Economic Affairs; Strumpel, B., Morgan J. N., Zahn, E., Eds.; Jossey-Bass, San Francisco, CA, USA.
(6) StataCorp. 2011. Stata Statistical Software, Release 12; StataCorp LP, College Station, TX, USA.
(7) Royston, P.; Altman, D. J. 1994. Regression using fractional polynomials of continuous covariates:
parsimonious parametric modeling (with discussion). Applied Statistics 43, 429-467.