東華大學圖書館 |

Valuation of Patents Using Stock Market Responses.

Record Type:	Electronic resources : Monograph/item
Title/Author:	Valuation of Patents Using Stock Market Responses./
Author:	Wang, Wenjun.
Published:	Ann Arbor : ProQuest Dissertations & Theses, : 2023,
Description:	108 p.
Notes:	Source: Dissertations Abstracts International, Volume: 85-03, Section: A.
Contained By:	Dissertations Abstracts International85-03A.
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30488874
ISBN:	9798380381017

Valuation of Patents Using Stock Market Responses.
Wang, Wenjun.

Valuation of Patents Using Stock Market Responses. - Ann Arbor : ProQuest Dissertations & Theses, 2023 - 108 p.

Source: Dissertations Abstracts International, Volume: 85-03, Section: A.

Thesis (Ph.D.)--University of California, Berkeley, 2023.

This item must not be sold to any third party vendors.

Patent valuation is of vital importance. This dissertation develops two patent valuation measures using abnormal stock market returns to patentee firms in time windows around grant and publication, respectively. The methodologies innovatively involve dynamics in the probability of grant, provide a way to estimate the patent value as early as the publication of the patent applications, and extend the valuation focus from patents to abandoned patent applications. The results provide important insights on the value of patents at publication and at grant, and the value of abandoned patent applications at publication. The results also shed light on the distribution of aggregate patent value, which has great potential to be used in R&D accounting. Moreover, the dissertation provides important and interesting findings about the relationship between patent value and grant lag. Furthermore, the dissertation also has important inferences on information flow in the patent application process at the USPTO.The measure of value at grant extends Kogan, Papanikolaou, Seru, and Stoffman (2017) (hereafter KPSS) to take advantage of information made available by the switch (for most publications) from publication at grant to publication eighteen months from filing. It uses a dynamic model on the changing probability of grant of a patent application as the lag from publication increases. Further, it focuses on sole-grant patents: the patents that are granted as the only patent for its assignee on the day of grant. The measure of value at publication is a unique contribution of this dissertation. I use it to measure the value of both patents and abandoned applications.The results show that if the dynamic decline in grant probability is ignored, as in KPSS, the average value of patents is overestimated by over 50%. The bias varies with the grant lag. However, with a dynamic probability of grant, the distributions of the value of patents with pre-grant publication and the value of patents for which the patentee opts to forego foreign applications in exchange for the right to delay publication until grant are similar. Moreover, with a dynamic probability of grant, the distributions of patent values at publication and patent values at grant are very similar. This implies that the market is risk-neutral and rational, and on average makes good patent value estimations at publication and grant.I use the estimated value to explore the relationship between patent value and the lag from publication. I find that (1) patent value at publication is not correlated with grant lag, indicating that the market participants cannot predict how long it will take for a patent application to get granted given the information available at the time of publication; (2) if the dynamic decline in grant probability is ignored (as in KPSS), the value of patents increases with the grant lag, (3) with a dynamic probability of grant, the value of patents does not tend to increase with grant lag. Indeed, the value tends to decrease modestly with lag from publication. In other words, more valuable patents tend to be granted somewhat earlier than less valuable ones.I explore the distribution of aggregate value and its potential for R&D accounting. I find that even if the estimate of value for each patent can have errors, according to the Central Limit Theorem, the aggregate value of patents can be quite accurate, with a narrow 95% confidence interval. Back-of-the-envelope calculations of aggregate patent value by year and aggregate patent value by CPC section are provided. This finding is important in providing crucial empirical verification for the validity of the KPSS model of patent valuation, as modified to include dynamic evolution of the probability of grant.The distribution of the value of patents and the value of abandoned applications at publication are similar, although the value of abandoned applications tends to be slightly lower than the value of granted patents. Most of the differences in distribution happen on the low-value part, while the high-value part of the distribution is very similar. This similarity still holds when controlling for the year of filing, CPC section, and assignee firm. The results suggest that the value of patent applications at publication is not importantly correlated with the probability of grant. This indicates that stock market participants cannot predict with accuracy at the time of publication whether a patent application will be granted or not. However, upon publication, they can make a virtually unbiased prediction of application value if granted. It also implies that when making decisions on patent grants, the USPTO focuses on patent validity, instead of patent value conditional on grant.The dissertation also has important inferences on information flow in the patent application process at the USPTO. I find that publication and grant are the two important events with information flow on patent value and probability of grant. At the time of publication, market participants form an initial estimation of patent value but cannot predict whether a patent application will be granted or not. At the time of grant, the market participants update the probability of grant to 100% and update their patent value estimation based on the final version of the patent and other available information. Before publication, there is little information flow because of the lack of information on the existence and the details of patent applications. After publication and before grant, the conditional probability of grant decrease with the time lag from publication but there is little information flow about patent value since the market participants cannot obtain access to real-time updates on patent application details without significant extra efforts or private information sources.

ISBN: 9798380381017Subjects--Index Terms:

Grant lag

Valuation of Patents Using Stock Market Responses.
LDR:07052nmm a2200397 4500 001 2394845
005 20240513061029.5
006 m o d
007 cr#unu||||||||
008 251215s2023 ||||||||||||||||| ||eng d
020 $a 9798380381017
035 $a (MiAaPQ)AAI30488874
035 $a AAI30488874
040 $a MiAaPQ $c MiAaPQ
100 1 $a Wang, Wenjun. $3 3281546
245 1 0 $a Valuation of Patents Using Stock Market Responses.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2023
300 $a 108 p.
500 $a Source: Dissertations Abstracts International, Volume: 85-03, Section: A.
500 $a Advisor: Wright, Brian.
502 $a Thesis (Ph.D.)--University of California, Berkeley, 2023.
506 $a This item must not be sold to any third party vendors.
520 $a Patent valuation is of vital importance. This dissertation develops two patent valuation measures using abnormal stock market returns to patentee firms in time windows around grant and publication, respectively. The methodologies innovatively involve dynamics in the probability of grant, provide a way to estimate the patent value as early as the publication of the patent applications, and extend the valuation focus from patents to abandoned patent applications. The results provide important insights on the value of patents at publication and at grant, and the value of abandoned patent applications at publication. The results also shed light on the distribution of aggregate patent value, which has great potential to be used in R&D accounting. Moreover, the dissertation provides important and interesting findings about the relationship between patent value and grant lag. Furthermore, the dissertation also has important inferences on information flow in the patent application process at the USPTO.The measure of value at grant extends Kogan, Papanikolaou, Seru, and Stoffman (2017) (hereafter KPSS) to take advantage of information made available by the switch (for most publications) from publication at grant to publication eighteen months from filing. It uses a dynamic model on the changing probability of grant of a patent application as the lag from publication increases. Further, it focuses on sole-grant patents: the patents that are granted as the only patent for its assignee on the day of grant. The measure of value at publication is a unique contribution of this dissertation. I use it to measure the value of both patents and abandoned applications.The results show that if the dynamic decline in grant probability is ignored, as in KPSS, the average value of patents is overestimated by over 50%. The bias varies with the grant lag. However, with a dynamic probability of grant, the distributions of the value of patents with pre-grant publication and the value of patents for which the patentee opts to forego foreign applications in exchange for the right to delay publication until grant are similar. Moreover, with a dynamic probability of grant, the distributions of patent values at publication and patent values at grant are very similar. This implies that the market is risk-neutral and rational, and on average makes good patent value estimations at publication and grant.I use the estimated value to explore the relationship between patent value and the lag from publication. I find that (1) patent value at publication is not correlated with grant lag, indicating that the market participants cannot predict how long it will take for a patent application to get granted given the information available at the time of publication; (2) if the dynamic decline in grant probability is ignored (as in KPSS), the value of patents increases with the grant lag, (3) with a dynamic probability of grant, the value of patents does not tend to increase with grant lag. Indeed, the value tends to decrease modestly with lag from publication. In other words, more valuable patents tend to be granted somewhat earlier than less valuable ones.I explore the distribution of aggregate value and its potential for R&D accounting. I find that even if the estimate of value for each patent can have errors, according to the Central Limit Theorem, the aggregate value of patents can be quite accurate, with a narrow 95% confidence interval. Back-of-the-envelope calculations of aggregate patent value by year and aggregate patent value by CPC section are provided. This finding is important in providing crucial empirical verification for the validity of the KPSS model of patent valuation, as modified to include dynamic evolution of the probability of grant.The distribution of the value of patents and the value of abandoned applications at publication are similar, although the value of abandoned applications tends to be slightly lower than the value of granted patents. Most of the differences in distribution happen on the low-value part, while the high-value part of the distribution is very similar. This similarity still holds when controlling for the year of filing, CPC section, and assignee firm. The results suggest that the value of patent applications at publication is not importantly correlated with the probability of grant. This indicates that stock market participants cannot predict with accuracy at the time of publication whether a patent application will be granted or not. However, upon publication, they can make a virtually unbiased prediction of application value if granted. It also implies that when making decisions on patent grants, the USPTO focuses on patent validity, instead of patent value conditional on grant.The dissertation also has important inferences on information flow in the patent application process at the USPTO. I find that publication and grant are the two important events with information flow on patent value and probability of grant. At the time of publication, market participants form an initial estimation of patent value but cannot predict whether a patent application will be granted or not. At the time of grant, the market participants update the probability of grant to 100% and update their patent value estimation based on the final version of the patent and other available information. Before publication, there is little information flow because of the lack of information on the existence and the details of patent applications. After publication and before grant, the conditional probability of grant decrease with the time lag from publication but there is little information flow about patent value since the market participants cannot obtain access to real-time updates on patent application details without significant extra efforts or private information sources.
590 $a School code: 0028.
653 $a Grant lag
653 $a Stock market
653 $a Patent valuation
653 $a Stock prices
653 $a Patent applications
690 $a 0338
690 $a 0503
690 $a 0438
710 2 $a University of California, Berkeley. $b Agricultural & Resource Economics. $3 1673761
773 0 $t Dissertations Abstracts International $g 85-03A.
790 $a 0028
791 $a Ph.D.
792 $a 2023
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=30488874