我有一个看起来像这样的 df （我只复制了数据集的一部分，其中包含从 1996 年 10 月 23 日到 2022 年 12 月 14 日的日期变量 %Y%m%d，其中缺少一些日子）。
PSA <- structure(list(date = structure(c(15657, 15665, 15671, 15672,
15678, 15679, 15685, 15686, 15693, 15714, 15721, 15728, 15735,
15742, 15749, 15763, 15770, 15777, 15779, 15780, 15784, 15785,
15789, 15790, 15791, 15797, 15798, 15804, 15805, 15810, 15811,
15812, 15819, 15819, 15824, 15825, 15831, 15832, 15832, 15833,
15839, 15839, 15840, 15840, 15841, 15846, 15846, 15847, 15848,
15853, 15854, 15855, 15859, 15860, 15860, 15861, 15861, 15862,
15867, 15867, 15868, 15868, 15869, 15874, 15874, 15875, 15875,
15876, 15880, 15881, 15881, 15882, 15882, 15883, 15888, 15888,
15889, 15889, 15890, 15894, 15895, 15895, 15896, 15896, 15897,
15901, 15902, 15902, 15903, 15903, 15904, 15905, 15909, 15909,
15910, 15911, 15912, 15915, 15916, 15916, 15917), class = "Date"),
country = c("Italy", "United Kingdom", "Italy", "United Kingdom",
"Italy", "United Kingdom", "Italy", "United Kingdom", "United Kingdom",
"United Kingdom", "United Kingdom", "United Kingdom", "United Kingdom",
"United Kingdom", "United Kingdom", "United Kingdom", "United Kingdom",
"United Kingdom", "Italy", "Italy", "United Kingdom", "Italy",
"Italy", "Italy", "Italy", "Italy", "Italy", "Italy", "Italy",
"United Kingdom", "Italy", "Italy", "Italy", "United Kingdom",
"Italy", "Italy", "Italy", "Italy", "Italy", "Italy", "Italy",
"Italy", "Italy", "United Kingdom", "Italy", "Italy", "Italy",
"Italy", "Italy", "Italy", "Italy", "Italy", "Italy", "Italy",
"Italy", "Italy", "United Kingdom", "Italy", "Italy", "Italy",
"Italy", "United Kingdom", "Italy", "Italy", "Italy", "Italy",
"United Kingdom", "Italy", "Italy", "Italy", "Italy", "Italy",
"United Kingdom", "Italy", "Italy", "Italy", "Italy", "United Kingdom",
"Italy", "Italy", "Italy", "Italy", "Italy", "United Kingdom",
"Italy", "Italy", "Italy", "Italy", "Italy", "United Kingdom",
"Italy", "Italy", "Italy", "Italy", "Italy", "Italy", "Italy",
"Italy", "Italy", "Italy", "Italy"), propIT_econ = c(20,
0, 12.5, 0, 12.5, 0, 25, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 5.26315789473684, 0, 0, 28.5714285714286, 0, 5.55555555555556,
23.5294117647059, 3.50877192982456, 21.0526315789474, 2.72727272727273,
0, 2.85714285714286, 0, 0.840336134453782, 0, 9.09090909090909,
14.5454545454545, 30.2325581395349, 13.5593220338983, 13.5593220338983,
2.85714285714286, 0, 0, 0, 0, 4.54545454545455, 5.61797752808989,
5.61797752808989, 0, 0, 0, 2.1505376344086, 4.8780487804878,
30, 3.2520325203252, 3.2520325203252, 3.77358490566038, 0,
3.06122448979592, 0, 2.01342281879195, 0, 0, 6.66666666666667,
4.65116279069767, 0, 0.975609756097561, 0, 1.53846153846154,
0, 11.6279069767442, 11.6279069767442, 5.14705882352941,
0, 1.61290322580645, 4.1958041958042, 4.1958041958042, 0.271739130434783,
0, 1.81818181818182, 7.2463768115942, 7.14285714285714, 0,
0, 0, 0, 13.5135135135135, 10.8108108108108, 10.8108108108108,
2.21606648199446, 0, 3.92156862745098, 8.57142857142857,
2.84090909090909, 2.84090909090909, 3.68421052631579, 9.61538461538462,
0, 1.44927536231884, 0.235849056603774, 0, 1.44230769230769
), propUK_econ = c(0, 0, 0, 1.51515151515152, 0, 0, 0, 0,
0, 0, 0, 0, 1.38888888888889, 1.53846153846154, 1.53846153846154,
2.85714285714286, 0, 0, 0, 0, 4.22535211267606, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0.840336134453782, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0)), row.names = c(NA, -101L), class = c("tbl_df",
"tbl", "data.frame"))
然后我将其转换为长格式：
PSAlong <- PSA %>% gather(variable, value, -date, -country)
我还有另一个 df，从 2004 年开始到 2022 年，有一个日期变量，而不是 %Y%qrt。
MIP_nat <- structure(list(date = c("2004.3", "2005.1", "2005.3", "2006.1",
"2006.3", "2007.1"), season = c("autumn", "spring", "autumn",
"spring", "autumn", "spring"), country = c("Italy", "Italy",
"Italy", "Italy", "Italy", "Italy"), Crime = c(23, 23, 25, 22,
25, 28), Public_transport = c(4, 5, 5, 0, 0, 0), Economic_situation = c(31,
37, 31, 33, 33, 28), Rising_prices_inflation = c(29, 32, 31,
28, 25, 26), Taxation = c(16, 14, 12, 15, 15, 18), Un_employment = c(31,
36, 31, 36, 30, 28), Terrorism = c(17, 7, 11, 7, 15, 9), Country_external_influence = c(0,
0, 0, 0, 0, 0), Government_debt = c(0, 0, 0, 0, 0, 0), Defence_Foreign_affairs = c(1,
1, 1, 1, 2, 1), Cyprus_issue = c(0, 0, 0, 0, 0, 0), Housing = c(1,
1, 1, 1, 3, 2), Immigration = c(17, 15, 15, 14, 20, 15), Health_social_security = c(5,
6, 6, 5, 6, 5), Education = c(3, 2, 4, 2, 1, 2), Pensions = c(9,
7, 10, 6, 9, 11), Environment_climate_energy = c(1, 2, 3, 2,
6, 9), International_situation = c(0, 0, 0, 0, 0, 0), Other = c(1,
1, 0, 0, 1, 1), None = c(0, 0, 0, 0, 0, 0), DN = c(0, 0, 1, 4,
0, 1)), row.names = c(NA, -6L), class = c("tbl_df", "tbl", "data.frame"
))
我尝试使用此代码将日期变量转换为日期，但返回 NA
MIP_nat$date <- as.yearqtr(MIP_nat$date)
我想要一个将第一个 df (PSAlong) 中的变量 "propIT_econ" 和 "propUK_econ" 的值与第二个 df (MIP_nat) 中的变量 "Economic_situation" 的值组合起来的图，或者具有相同的值具有不同比例的两个 x_轴的比例。 我很挣扎，因为日期变量有两种不同的格式，这使得我只能拥有 df 的变量，而不是两者。例如，这工作正常： 我只用这段代码运行第一个图，它有效
PSAlong %>% filter(variable == "propIT_econ" | variable == "propUK_econ") %>%
ggplot(aes(x = as.Date(date), y = value, color = variable)) +
geom_step(linewidth = 0.7, alpha = 0.8) +
scale_x_date(date_breaks = "1 year",
date_labels = "%Y")
我尝试过，但它没有产生我想要的结果：
ggplot() +
geom_step(data = PSAlong %>% filter(variable == "propIT_econ" | variable == "propUK_econ),
aes(x = as.numeric(date), y = value, color = country)) +
geom_line(data = MIP_nat,
aes(x = as.numeric(date), y = "Economic_situation", linetype = country)) +
scale_x_continuous(limits = c(1996,2023),
breaks = seq(1997,2023, by = 1))
谢谢您的帮助！