International Journal of Technical Research and Applications e-ISSN: 2320-8163,

www.ijtra.com Volume-2, Special Issue 4 (July-Aug 2014), PP. 08-12

8 | P a g e
DETERMINANTS OF PATENT VALUE IN US
AND JAPANESE UNIVERSITY PATENTS
Kallaya Tantiyaswasdikul
Graduate School of Policy Science
Ritsumeikan University
Kyoto, Japan

Abstract This study investigates the determinants of patent
value in US and Japanese university patents. Four broad value
determinants including the technical background of patents, the
distance of technology from the application date to present, the
breadth or scope of patent protection, and the technology
classification are considered. The key variables are the number of
forward citations as a dependent variable and independent
variables composed of the number of backward citations, years,
claims, and IPC classes. A comparison analysis using zero-
inflated negative binomial regression between US and Japanese
university patents has been proved based on the notion of the
determinants of patent value. The results reveal that both the US
and Japanese university patents share common determinants of
value. Older patents receive more citations than younger patents.
Backward citations and claims have positive and significant
impact on the patent value; however, IPC classes reflect no
impact on the value of patents.
I ndex Terms Japanese university patents, patent value,
university patents, US university patents.
I. INTRODUCTION
Radical changes have been observed in the academic
patenting behavior over the past 30 years since the enactment
of the Bayh-Dole Act in the US in 1980 that allows universities
to retain their rights in any inventions deriving from public-
funded research. For Japan, these changes have impacted the
Japanese university patenting due to the emulation of the Act in
1999. Japan has adopted a Bayh-Dole Act-like model to
enhance the effectiveness of university-industry technology
transfer. As a consequence of this change, a higher propensity
to patent academic inventions has been observed by Motohashi
and Muramatsu [1]. At the same time, scholars and
policymakers have underlined the crucial role played by
industryuniversity partnerships in the knowledge society as
discussed by Etzkowitz [2], Jaffe [3], Mansfield [4], [5],
Mansfield and Lee [6], and Meyer [7].
Although recent studies of Guellec and van Pottelsberghe
de la Potterie [8], Mansfield [5], McMillan et al. [9], and Narin
et al. [10] clearly illustrate the significant contribution of public
research (performed in universities and public laboratories) to
the innovation performances of the business sector, mainly
through knowledge spillovers, the rapid increase in academic
patenting has provoked new debates about the quality of these
patents. Do they herald a surge in academic inventions, or do
they merely reflect a higher propensity to patent inventions of
lower quality?
This study intends to contribute to this debate. It aims to
investigate the value of academic patents and compare their
value determinants in US and Japanese university patents, as
well as Japanese university patents with different assignees. In
order to determine the value of patents, various measures have
been developed. All of these measures can be derived from
patent data directly. I consider four broad value determinants
including the technical background of patents, the distance of
technology from the application date to present, the breadth or
scope of patent protection, and the technology classification.
The key variables are the number of forward citations as a
dependent variable and independent variables composed of the
number of backward citations, years, claims, and international
patent classification (IPC) classes.
In this study, a comparison analysis using zero-inflated
negative binomial (ZINB) regression between US and Japanese
university patents has been proved based on the notion of the
determinants of patent value. Additionally, I provide detailed
analyses of Japanese university patents with different
institutional-type settings, including university assignee and
university co-assignee or university-industry collaboration
(UIC) patents. The results reveal that both the US and Japanese
university patents share common determinants of value. More
precisely, in the case of Japanese university patents, the
evidence suggests that the breadth of patent protection (claims)
significantly affects valuations, but there is a difference in
terms of the nature of patents between university assignee and
UIC patents. The remainder of this article is structured as
follows: The summary of the determinants of patent value is
presented in section 2. Section 3 explains the methods of this
study. Section 4 provides data collection and data set, while
Section 5 presents the empirical analysis and findings. The last
section gives conclusions.
II. DETERMINANTS OF PATENT VALUE
According to the study of Stevens and Burley [11], it is
known that on average only one to three patents out of 100
yield significant financial return. This skewed distribution of
patent value has been at the origin of a small but growing
stream of economic research that attempts to identify the
determinants of patent value as discussed by Griliches [12],
Griliches et al. [13], Pakes [14], Pakes and Schankerman [15],
Sapsalis et al. [16], Scherer [17], and Scherer and Harhoff
[18].
Regarding the skew distribution of patent value, additional
information that correlated with the value of patent rights has
been employed to estimate the valuation. Various indicators
have been used as variables to determine patent value in the
economic literature on the measurement of the value of
patents, such as the number of times the patent is cited
(forward citations), or the length of its renewal, or the number
of countries where it is taken (patent family size), or the
breadth or scope of patent protection (patent claims).
Pakes [14], Pakes and Schankerman [15], and
Schankerman and Pakes [19] were the first to develop and
estimate models in which the observed renewal decisions are
used to estimate the distribution of patent values. Trajtenberg
[20] computed a measure of social returns to the computer-
tomography scanner industry and relates that measure to
citation indicators. Lerner [21] examined the impact of patent
scope on the market value of biotechnology firms and
developed a proxy for the breadth of patent protection to
International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume-2, Special Issue 4 (July-Aug 2014), PP. 08-12
9 | P a g e
determine the valuation. Putnam [22] integrated application
data into the analysis of the value of patent.
Harhoff et al. [23] estimated patent value using a broad set
of indicators, which are composed of the number of citation
received from subsequent applications, the number of
references to prior patents (backward citations), the number of
references made to the non-patent literature, the outcome of
opposition proceedings, the patent family size, and the number
of different four-digit IPC classifications. They found
significant correlations between patent value and citations
received from subsequent patents as well as backward
citations. They also found that the observed outcomes of
opposition cases and the measure of international patent
families are particularly valuable.
Sapsalis et al. [16] compared corporate and academic
patents to assess whether they have similar value distributions
and share common determinants of value. To evaluate the
value determinants of patents, they used the number of non-
patent citations, backward citations, co-assignees, and
members in the patent family as indicators. They found that
the value distribution of academic patents is very close to that
for corporate patents and the determinants of patent value are
broadly similar for the two sectors. Backward citations, non-
patent citations, and the number of inventors and co-assignees
all affect the value of both academic and corporate patents.
Guellec and van Pottelsberghe de la Potterie [24] observed
the probability of getting a patent granted to approximate the
value of a patent. They used the indicators of the patenting
strategy, the domestic and international R&D collaboration,
the technological diversity (the number of IPC classes), and
the mix of designed states for protection (the patent family
size) to determine the value of patents. They found that the
strategic decision provides the useful information about the
grant probability, while the technical diversity has a negative
impact on the probability of grant, and the link between patent
value and family size is ambiguity.
The type and number of explanatory variables that have
been used as determinants of patent value vary widely across
studies. The most frequently used determinants are the number
of forward citations (when it is not used as a dependent
variable), the number of backward citations, and the
geographical scope for protection (the number of countries in
the patent family). Other variables rely on the concepts of
opposition procedures, renewal data, application scope (the
number of claims), and non-patent citations.
III. METHODOLOGY
This study examines the determinants of patent value in US
and Japanese university patents using panel data from the
United States Patent and Trademark Office (USTPO). For this
purpose, multiple regression through zero-inflated negative
binomial (ZINB) regression was employed. In most patent
applications, the value of the forward citation, which is used as
the response variable, is zero in a large number of patents. This
fact may make it difficult to successfully apply linear
regression to the data. Thus, this study is based on a logistic
model, wherein patents whose forward citation is equal to or
beyond a certain threshold can be differentiated from others.
ZINB models, which are robust against over-dispersion caused
by a large number of zero counts, are used in bibliometric
studies, including patent analyses as presented by Foltz et al.
[25], Lee et al. [26], Tang and Shapira [27], and Yoshikane
[28].
ZINB regression with the response variable is the number
of forward citations was employed. This is the most
commonly used proxy for the value of patents. For the
explanatory variables, I develop four broad value determinants
to determine patent value as follows: (1) the technical
background of a patent (measured by the number of backward
citations); (2) the distance of technology from the application
date to present (measured by the year filed of patents); (3) the
breadth or scope of patent protection (measured by the
number of claims); (4) the technology classification (measure
by the number of IPC classes).
Despite the heterogeneity of previous studies, some
similarities emerge. The most important is probably the fact
that the number of forward patent citations is closely
associated with the value of a patent; all studies using forward
patent citations reach this conclusion as discussed by Sapsalis
et al. [16]. Thus, I use forward citations represent patent value
and estimate their value determinant through four independent
variables including backward citations, years, IPC classes, and
claims.
Future citations received by a patent (forward citations) are
one indication that an innovation has contributed to the
development of subsequent inventions. For this reason,
citations have been used as a measure of the value of an
invention as explored by Trajtenberg [20]. An inventor must
cite all related prior US patents in the patent application. A
patent examiner who is an expert in the field is responsible for
insuring that all appropriate patents have been cited. Like
claims, the citations in the patent document help to define the
property rights of the patentee as examined by Lanjouw and
Schankerman [29].
The number of claims is another, underutilized, indicator
of the bits of information contained in a patent, and therefore
of its value. Supporting evidence for the relationship between
claims and value is found in the fact that claims are positively
correlated with forward and backward citations as investigated
by Lanjouw and Schankerman [30]. Tong and Frame [31]
suggested that patent claims might be a better indicator of
technological effort than straight patent counts. Certainly,
claims correlate better with other technology-related indicators
than patent counts.
For technology classification, the use of classifications
helps to expedite prior art searches, and helps avoid possible
ambiguity that may be present in other keyword search fields
as discussed by Harris et al. [32]. The IPC system divides
technology into eight discrete sections, including section A:
Human necessity; section B: Performing operations,
Transporting; section C: Chemistry, Metallurgy; section D:
Textiles, Paper; section E: Fixed constructions; section F:
Mechanical engineering, Lighting, Heating, Weapons,
Blasting; section G: Physics, and section H: Electricity.
The IPC is a technology-based classification system with
approximately 70,000 subdivisions. According to Adams [33]
and Tantiyaswasdikul [34], in practice, there are few
inventions that can be classified into one particular
technology; most of the innovations include hybrid elements
and patents may be assigned to more than one subclass. Like
Lerner [21] and Tantiyaswasdikul [34], I use the set of all 4-
digit IPC subclasses to which each patent was assigned for this
analysis.

International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume-2, Special Issue 4 (July-Aug 2014), PP. 08-12
10 | P a g e
IV. DATA COLLECTION AND DATA SET
This study focuses on the analysis of the determinants of
patent value by providing a comparison between US and
Japanese university patents. Additionally, the study also
investigates in detail Japanese university patents with different
institutional types of assignees. Since the Japan UIC policy
initiatives that have been implemented since 1998 and the
number of Japanese university patents owned by universities
are limited by institutional and regulatory disincentives, the
data in this analysis cover the period after the enactment of UIC
policies from 1998 to 2008. For Japan, the data set consists of
all Japanese national university-granted US Utility Patents that
were applied from 1998 to 2008. For the US, the data set is
composed of a 5% random sample of the US university-granted
US Utility Patents that were applied between 1998 and 2008.
I employ ZINB regression where the dependent variable is
the number of forward citations. The explanatory variables
include the number of IPC classes, the number of claims, the
number of years filed, and the number of backward citations.
Table I presents the descriptive statistics of variables of US and
Japanese university patents and Table II presents the
descriptive statistics of variables of Japanese university
assignee and UIC patents.
TABLE I. DESCRIPTIVE STATISTIC OF VARIABLES OF US AND
JAPANESE UNIVERSITY PATENTS
Variables Descriptive statistic

Obs. Mean Std. Dev. Max Min
US university patents
Forward citations
1,755 1.98 4.36 40 0
IPC classes 1,755 1.68 0.98 9 1
Claims 1,755 20.61 16.56 127 1
Years 1,755 9.18 3.19 15 5
Backward citations 1,755 11.99 15.60 98 0
JP university patents
Forward citations
1,779 0.98 2.85 46 0
IPC classes 1,779 1.51 0.84 7 1
Claims 1,779 11.56 7.88 100 0
Years 1,779 7.50 2.52 15 5
Backward citations 1,779 5.41 6.05 123 0
Note: All data were obtained from the online records system of USPTO website;
http://www.uspto.gov/patents/process/search/
(Update 20 November 2013).

The count number of US university patents is 1,755 and the
count number of Japanese university patents is 1,779,
respectively. Additionally, in the case of Japan, the patents
were classified according to assignees and institution type.
Specifically, the patents were divided into university assignee
patents and university co-assignee or UIC patents. The number
of university assignee patents is 916, while the number of UIC
patents is 863.
TABLE II. DESCRIPTIVE STATISTIC OF VARIABLES OF JAPANESE
UNIVERSITY ASSIGNEE AND UIC PATENTS
Variables Descriptive statistic

Obs. Mean Std. Dev. Max Min
US university patents
Forward citations
916 1.17 3.04 38 0
IPC classes 916 1.59 0.88 7 1
Claims 916 11.05 7.59 57 0
Years 916 8.31 2.86 15 5
Backward citations 916 4.95 6.64 123 0
JP university patents
Forward citations
863 0.78 2.63 46 0
IPC classes 863 1.42 0.78 6 1
Claims 863 12.09 8.15 100 1
Years 863 6.65 1.73 15 5
Backward citations 863 5.90 5.32 36 0
Note: All data were obtained from the online records system of USPTO website;
http://www.uspto.gov/patents/process/search/
(Update 20 November 2013).
V. EMPIRICAL ANALYSIS AND FINDINGS
The results of ZINB regression analyses are presented in
Tables III and IV. The results reveal that patent values for US
and Japanese university patents seem to react to almost similar
determinants. Older patents receive more citations than
younger patents. Backward citations have positive and
significant impact on the number of forward citations. Claims
have positive impact to patent value; however, it reveals
significance only on US university patents. IPC classes have no
impact on the number of forward citations, as demonstrated in
Table III.
However, the result of Japanese university patents in Table
III is an aggregate number of patents that combine both the
university assignee and UIC patents. To investigate why the
measure of claims has no impact on Japanese university
patents, while this factor has a positive and significant impact
on US university patents, a detailed analysis of determinants
of patent value in Japanese university patents was created.
Table IV provides the results of a comparison analysis
between Japanese university assignee and UIC patents.
We can observe almost similar results as the comparison of
value determinants between US and Japanese university patents
in Table III, except that the impact of claims on the patent
value is different between university assignee and UIC patents.
In the case of university assignee patents, the number of claims
has a significantly positive impact on patent value but the
number of claims has no impact on patent value in the case of
UIC patents.
TABLE III. REGRESSION COEFFICIENTS FOR RESPONSE VARIABLE:
NUMBER OF FORWARD CITATIONS FOR US AND JAPANESE UNIVERSITY
PATENTS
Variables Regression Coefficients

US University patents J P university patents
IPC classes -0.005
(0.034)
-0.071
(0.043)
Claims 0.007***
(0.002)
0.005
(0.004)
Years 0.135***
(0.013)
0.159***
(0.013)
Backward citations 0.011***
(0.002)
0.018***
(0.006)
Obs. 1755 1779
Constant -0.315 -0.453
Log likelihood -1812.669 -1186.419
LR chi 2(4) 178.11 146.47
Note: ***represent statistical significance at the 1% level; standard errors in parentheses.


TABLE IV. REGRESSION COEFFICIENTS FOR RESPONSE VARIABLE:
NUMBER OF FORWARD CITATIONS FOR JAPANESE UNIVERSITY ASSIGNEE
AND UIC PATENTS
Variables Regression Coefficients

University assignee patents UI C patents
IPC classes -0.049
(0.054)
-0.083
(0.073)
Claims 0.011*
(0.006)
-0.003
(0.005)
Years 0.134***
(0.019)
0.213***
(0.023)
Backward citations 0.013*
(0.007)
0.023*
(0.010)
Obs. 916 863
Constant -0.301 -0.773
Log likelihood -685.6705 -494.8158
LR chi 2(4) 58.15 86.30
Note: ***, * represent statistical significance at the 1% and 10% levels; standard errors in parentheses.
It is interesting that when analysis is performed separately
between Japanese university assignee and UIC patents, we can
observe the result of the impact of claims on the patent value
in the case of university assignee patents, which show the
International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume-2, Special Issue 4 (July-Aug 2014), PP. 08-12
11 | P a g e
similarity to the US university patents, which show positive
significance. On the other hand, this is opposite to the case of
UIC patents, when the number of claims has no impact to
patents value.
The results of this analysis correspond to the existing study
of Harhoff et al. [23], Lanjouw and Schankerman [30], and
Tantiyaswasdikul [34] where a number of indicators are
significantly correlated with patent value. The measure for
references to the patent literature or backward citations carries
significant positive coefficients to patent value, similar to the
evidence in the study of Harhoff et al. [23]. Likewise, the
number of years filed has significantly positive impact to
patent value.
The claims number is a particularly good predictor of
patent value when it reveals a positive correlation with the
increased number of forward citations. Supporting evidence
for the relationship between claims and value is found in the
fact that claims are positively correlated with forward citations
in the study of Lanjouw and Schankerman [30] and
Tantiyaswasdikul [34].
Contrary to the previous results of Lerner [21], I find that
the number of four-digit IPC classifications has a negative
impact on the patent value. However, the relationship between
the indicator of IPC class and patent value is ambiguous since
Harhoff et al. [23] found that the number of four-digit IPC
classifications does not have any explanatory power. The
ambiguity of IPC class as a variable in the study of patent
value is found in the analysis of Guellec and van Pottelsberghe
de la Potterie [24] when the technical diversity has a negative
impact on the probability of patent grant. The higher the
number of IPC classes listed in an application, the lower the
chance to get a grant. The explanation is due to the fact that it
is possible that a high number of classes may reflect not only
the technological diversity of the invention, but also the
perplexity of the examiner facing a somewhat unclear
technology as discussed by Guellec and van Pottelsberghe de
la Potterie [24].
The explanation of the relationship of the breadth of patent
protection and patent value can be explained regarding the
links between an innovation and its technological antecedents
and descendants. For claims, the number of claims in an
existing patent has some relation to the technological
innovation of previous patents. According to Tong and Frame
[31], patents do not measure fundamental units of
inventiveness. This privilege lies in the domain of patent
claims. Thus, an inventors invention is embodied in his or her
claims.
A new invention based on the existing notion will have a
few claims since the knowledge of that invention relates to the
antecedent technology. In contrast, for the new discovery, the
number of claims tends to be excessive. For the explanation of
the different impact of claims on university assignee and UIC
patents consider the following. In the case of UIC patents,
when the number of claims increases the number of forward
citations decreases. In general, when the number of claims
increases the value of patents decreases.
Since UIC patents are the results of the collaborative
research between university and industry that have
commercialization purposes, the number of claims can cause
difficulty of accessibility of invention in the future. Thus, a
small number of claims is better for the broader targets. In
contrast, in the case of university assignee patents, when the
number of claims increases the number of forward citations
also increases. This result corresponds to the existing study of
Lanjouw and Schankerman [30] and Tantiyaswasdikul [34]
when the number of claims reflects the value of patents.
Important inventions gain many citations received. Moreover,
the number of claims reflects freshness that means new
inventions provide some incentives to researchers and the
researcher would like to catch up new technology. Considering
this point when the number of claims increases, the number of
forward citations also increases.
For IPC classes representing the technology fields,
according to Lerner [21], the number of IPC classes has a
positive impact on the number of forward citations. This is
understandable since a patent that falls into many technology
fields provides many possibilities for researchers in many
areas to cite. In this case, the number of IPC classes indicates
the quantity aspect. However, in this result, when the number
of IPC classes decreases, the number of forward citations
increases, so the number of IPC classes indicates the quality
aspect. This result corresponds to the existing study of
Tantiyaswasdikul [34]. The explanation is that it is possible
that a high number of classes may reflect not only the
technological diversity of the invention, but also the perplexity
of the unclear technology as discussed by Guellec and van
Pottelsberghe de la Potterie [24].
The measure for references to the patent literature or
backward citations carries significant positive coefficients to
patent value similar to the evidence in the study of Harhoff et
al. [23]. This evidence reflects the relationship between
technological antecedents and descendants or backward and
forward citations of innovation when an invention based on an
existing technology represents the important innovation as
discussed by Trajtenberg et al. [35].
VI. CONCLUSION
This study has been an attempt to use information from
patent applications to determine patent value. The analysis of
these data has been quite promising. Clear evidence of the
significant correlation between the provided indicators and
patent value has been observed. The results reveal that patent
value for US and Japanese universities seems to react to
almost similar determinants. Older patents receive more
citations than younger patents. Backward citations and claims
have positive and significant impact on the patent value;
however, IPC classes reflect no impact on the value of patents.
Moreover, regarding the breadth of patent protection in terms
of claims, the results reveal the difference between Japanese
university assignee and UIC patents. In the case of university
assignee patents, the number of claims has a significantly
positive impact on patent value but the number of claims has
no impact on patent value in the case of UIC patents.
In light of the findings of this study, considering the
information derived from patent data is important since it
provides not only the technological antecedents and
descendants of innovation, but also the determinant of patent
value. Information on the value of a patent is contained not
only in forward citations as recognized in previous studies, but
also in other variables such as the technical background of
patents and the breadth or scope of patent protection.
ACKNOWLEDGMENT
I would like to express my gratitude and appreciation to
Professor Oda Hisaya, Graduate School of Policy Science,
Ritsumeikan University for his support and encouragement.
International Journal of Technical Research and Applications e-ISSN: 2320-8163,
www.ijtra.com Volume-2, Special Issue 4 (July-Aug 2014), PP. 08-12
12 | P a g e
REFERENCES
[1] K. Motohashi and S. Muramatsu, Examining the university-
industry collaboration policy in Japan: Patent analysis,
Technology in Society, vol. 34(2), pp.149162, 2012.
[2] H. Etzkowitz, The norms of entrepreneurial science: Cognitive
effects of the new university-industry linkages, Research
Policy, vol. 27(8), pp. 823833, 1998.
[3] A.B. Jaffe, Real effects of academic research, The American
Economic Review, vol. 79(5), pp. 957970, 1989.
[4] E. Mansfield, Academic research and industrial innovation,
Research Policy, vol. 20(1), pp. 112, 1991.
[5] E. Mansfield, Academic research and industrial innovation: An
update of empirical findings, Research Policy, vol. 26(7), pp.
733776, 1998.
[6] E. Mansfield and J.Y. Lee, The modern university: Contributor
to industrial innovation and recipient of industrial R&D
support, Research Policy, vol. 25(7), pp. 10471058, 1996.
[7] M. Meyer, Academic entrepreneurs or entrepreneurial
academics? Research-based ventures and public support
mechanisms, R&D Management, vol. 33(2), pp. 107115,
2003.
[8] D. Guellec and B. Van Pottelsberghe De La Potterie, From
R&D to productivity growth: Do the institutional settings and
the source of funds of R&D matter?, Oxford Bulletin of
Economics and Statistics, vol. 66(3), pp. 353378, 2004.
[9] G.S. Mcmilan, F. Narin, and D.L. Deeds, An analysis of the
critical role of public science in innovation: The case of
biotechnology, Research Policy, vol. 29(1), pp. 18, 2000.
[10] F. Narin, K.S. Hamilton, and D. Olivastro, The increasing
linkage between US technology and public science, Research
Policy, vol. 26(3), pp. 317330, 1997.
[11] G.A. Stevens and J. Burley, 3,000 raw ideas=1 commercial
success, Research Technology Management, vol. 40(3), pp. 16
27, 1997.
[12] Z. Griliches, Patent statistic as economic indicators: A survey,
Journal of Economic Literature, vol. 28(4), pp. 16611707,
1990.
[13] Z. Griliches, A. Pakes, and B.H. Hall, The value of patents as
indicators of inventive activity, Working Paper no. 2083,
NBER, Cambridge, MA, 1986.
[14] A. Pakes, Patents as options: Some estimates of the value of
holding European patent stocks, Ecomometrica, vol.54, pp.
755784, 1986.
[15] A. Pakes and M. Schankerman, The rate of obsolescence of
patents, research gestation lags, and the private rate of return to
research resources, in R&D, Patents, and Productivity, Z.
Griliches, Ed. Chicago: University of Chicago Press, 1984, pp.
7388.
[16] E. Sapsalis, B. Van Pottelsberghe De La Potterie, R. Navon,
Academic versus industry patenting: An in-depth analysis of
what determines patent value, Research Policy, vol. 35(10), pp.
16311645, 2006.
[17] F.M. Scherer, Firm size, market structure, opportunity, and the
output of patented inventions, The American Economic
Review, vol. 55, pp. 10971125, 1965.
[18] F.M. Scherer and D. Harhoff, Technology policy for a world of
skew-distributed outcomes, Research Policy, 29(40), pp. 559
566, 2000.
[19] M. Schankerman and A. Pakes, Estimates of the value of patent
rights in European countries during the post-1950 period,
Economic Journal, vol. 97, pp. 125, 1986.
[20] M. Trajtenberg, A penny for your quotes: Patent citations and
the value of inventions, RAND Journal of Economics, vol. 21,
pp. 172187, 1990.
[21] J. Lerner, The importance of patent scope: An empirical
analysis, RAND Journal of Economics, vol. 25(2), pp. 319
333, 1994.
[22] J.D. Putnam, The value of international patent rights, Ph.D.
diss., Yale University, 1996.
[23] D. Harhoff, F.M. Scherer, and K. Vopel, Citations, family size,
opposition and the value of patent rights, Research Policy, vol.
32(8), pp. 13431363, 2002.
[24] D. Guellec and B. Van Pottelsberghe De La Potterie,
Applications, grants and the value of patent, Economics
Letters, vol. 69(1), pp. 109114, 2000.
[25] J. Foltz, B. Barham, and K. Kim, Universities and agricultural
biotechnology patent production, Agribusiness, vol. 16(1), pp.
8295, 2000.
[26] Y.G. Lee, J.D. Lee, Y.I. Song, and S.J. Lee, An in-depth
empirical analysis of patent citation counts using zero-inflated
count data model: The case of KIST, Scientometrics, vol. 70(1),
pp. 2739, 2007.
[27] L. Tang and P. Shapira, Effects of international collaboration
and knowledge moderation on Chinas nanotechnology research
impacts, Journal of Technology Management in China, vol.
7(1), pp. 94110, 2012.
[28] F. Yoshikane, Multiple regression analysis of a patents citation
frequency and quantitative characteristics: The case of Japanese
patents, Scientometrics, vol. 96(1), pp. 365379, 2013.
[29] J.O. Lanjouw and M. Schankerman, Characteristics of patent
litigation: A window on competition, RAND Journal of
Economics, vol. 32(1), pp. 129151, 2001.
[30] J.O. Lanjouw and M. Schankerman, The quality of ideas:
Measuring innovation with multiple indicators, Working Paper
no. 7345, NBER, Cambridge, MA, 1999.
[31] X. Tong and J.D. Frame, Measuring national technological
performance with patent claims data, Research Policy, vol.
23(2), pp. 133141, 1994.
[32] C.G. Harris, R. Arens, and P. Srinivasan, Comparison of IPC
and USPC classification systems in patent prior art searches,
Proceedings of the 3
rd
international workshop on patent
information retrieval, ACM: 2732.
[33] S. Adam, Comparing the IPC and the US classification systems
for the patent searchers, World Patent Information, vol. 23(1),
pp. 1523, 2001.
[34] K. Tantiyaswasdikul, The impact of the breadth of patent
protection and the Japanese university patents, International
Journal of Innovation, Management and Technology, vol. 3(6),
pp. 754758, 2012.
[35] M. Trajtenberg, R. Henderson, and A.B. Jaffe, University
versus corporate patents: A window on the basicness of
invention, Economics of Innovation and New Technology, vol.
5(1), pp. 1950, 1997.